Upload
others
View
4
Download
0
Embed Size (px)
Citation preview
Management and nutritional
strategies to improve the postnatal
performance of light weight pigs
By
Sadie Louise Douglas
BSc (Hons) Bioveterinary Science
MSc (Hons) Control of infectious disease in animals
A thesis submitted for the degree of Doctor of Philosophy
(PhD)
School of Agriculture, Food and Rural Development
Newcastle University
Submitted February 2014
i
ABSTRACT
During the production period from birth to slaughter there are some pigs that grow
markedly slower, despite conditions that seem to support the rapid growth of their
contemporaries. This reduction in growth inevitably leads to weight variation within a
group and results in system inefficiencies. The aim of this thesis was to identify risk
factors involved in poor growth and to develop management and nutritional treatments
to enable light pigs to maximise their growth at different stages of production.
Risk factor analysis for a large dataset showed that, in particular, low birth and weaning
weight result in poor growth to finishing. Some light pigs do, however, have the
capacity to compensate for low weight at earlier stages of production. Preweaning
intervention demonstrated that low birth weight pigs cross fostered into litters with
similar weight littermates had a significantly higher weaning weight than those in mixed
litters with heavier pigs; however the provision of supplementary milk to such litters
had no further beneficial effect. A post weaning feeding regime formulated for low birth
weight pigs, with a higher nutrient specification diet based on more digestible
ingredients, not only showed improved performance to 10 weeks of age, but also
enabled low birth weight pigs to meet the BW of heavier birth weight pigs. In contrast,
a high specification diet (higher in amino acid: energy content) had no effect on the
growth of low birth weight pigs when offered from 9 weeks of age, suggesting a critical
window for intervention.
Overall, the crucial stages of postnatal growth for light pigs have been identified, and
preweaning and early post weaning treatments have been developed. These not only
improve the performance of low birth weight pigs but also allow them to catch up with
heavier birth weight pigs.
ii
Declaration
This thesis has been composed by myself and has not been submitted as part of any
previous application for a degree. All sources of information have been specifically
acknowledged by means of referencing.
Sadie Louise Douglas
iii
Acknowledgements
Firstly I would like to thank the British Pig Executive for sponsoring my PhD and the
support and guidance they have provided throughout.
A huge thank you goes to my supervisors Professor Ilias Kyriazakis and Professor
Sandra Edwards for the encouragement and guidance they have provided over the
course of my PhD. I feel incredibly lucky to have had two supervisors with such
expertise, who despite their busy schedules, always found time to give me support when
it was needed.
Completing my experimental work would not have been possible without the help of all
the staff at Cockle Park Farm, who always accommodated me despite my sometimes
difficult requests. In particular, Darren, Mark and Jim who not only provided invaluable
knowledge and assistance but also kept me entertained during the weekends and
holidays, making my time on farm so enjoyable. Also to Paul Lamothe for his
invaluable help with my behavioural data analysis during his stay in Newcastle. I would
also like to thank all the staff at Newcastle University who has provided assistance at
various points during my PhD, in particular Peter Avery for his statistical expertise,
Steve Hall for his never ending IT support and Debra Patterson for her impressive
organisational skills.
A special thank you goes to Ian Wellock and Primary Diets for their invaluable help in
diet formulation and for their advice and expertise. I would also like to thank ACMC
and Genus for kindly providing data.
I cannot thank my parents enough for their encouragement and support throughout my
PhD as well as the rest of my education, they have provided me with so many
opportunities that would have not been possible without them. And finally to Mathew,
for his understanding of my absence at times, unwavering support of my goals and for
his continued support.
iv
Whilst I have not been able to thank everyone personally, I now realise how much work
and assistance is required to complete a PhD, and I am extremely grateful to anyone
who has helped me at any point during the completion of this thesis.
v
Publications and conference abstracts
Peer-reviewed publications
Douglas, S. L., S. A. Edwards, E. Sutcliffe, P. W. Knap, and I. Kyriazakis. 2013.
Identification of risk factors associated with poor lifetime growth performance in pigs.
Journal of Animal Science. 91: 4123-4132.
Douglas, S. L., S. A. Edwards and I. Kyriazakis. 2014. Management strategies to
improve the performance of low birth weight pigs to weaning and their long term
consequences. Journal of Animal Science. 92: 2280-2288.
Douglas, S. L., S. A. Edwards, I.J. Wellock and I. Kyriazakis. Under review. High
specification starter diets improve the performance of low birth weight pigs to 10 weeks
of age. Journal of Animal Science.
Conference and abstracts
Douglas, S.L., S. A. Edwards, and I. Kyriazakis. (2013). Can low birth weight exhibit
catch up growth post weaning if fed according to size? Proceedings for the British
Society of Animal Science, 16-17th
April 2013. Theatre Presentation.
Douglas, S.L., S. A. Edwards, and I. Kyriazakis. (2013). Can low birth weight pigs
exhibit catch up growth post weaning if fed according to size? European Symposium of
Porcine Health Management, Edinburgh, 22-24th
May 2013. Poster presentation.
Douglas, S.L., S. A. Edwards, and I. Kyriazakis. (2012). Identification of risk factors
associated with poor growth performance in pig. Proceedings for the British Society of
Animal Science, 24-25th
April 2012. Theatre Presentation.
vi
List of abbreviations
ADG Daily live weight gain
ADFI Average daily feed intake
ADMI Average daily milk intake
BiW Birth weight
BW Body weight
CP Crude Protein
CRL Crown rump length
CV Coefficient of variation
d Day
DE Digestible Energy
FW Final weight
H Hour
FCE Feed conversion efficiency
FI Feed intake
IW Intermediate body weight
kg Kilogram
LBiW Low birth weight
m Metre
mo Month
NBiW Normal birth weight
NE Net energy
OR Odds ratio
PI Ponderal index
SD Standard deviation
SDG Scaled ADG
SFI Scaled FI
vii
wk Week
WW Weaning weight
viii
Contents
Chapter 1. Introduction ..................................................................................................... 1
1.1 What are light pigs? ................................................................................................ 1
1.2 Problems associated with light pigs ........................................................................ 2
1.3 How has the industry attempted to deal with light pigs? ........................................ 3
1.4 Thesis aims .............................................................................................................. 5
Chapter 2. The risk factors associated with poor growth performance in pigs ................. 6
2.1 Animal characteristics ............................................................................................. 6
2.2 The prenatal environment ....................................................................................... 7
2.2.1 Intrauterine growth restriction.......................................................................... 8
2.2.2 Birth weight ...................................................................................................... 9
2.3 The postnatal environment .................................................................................... 10
2.3.1 Lactation ......................................................................................................... 10
2.3.2 Weaning ......................................................................................................... 12
2.3.3 Post weaning .................................................................................................. 13
2.4 Conclusions ........................................................................................................... 15
Chapter 3: Identification of risk factors associated with poor lifetime growth
performance in pigs ......................................................................................................... 17
3.1 Introduction ........................................................................................................... 17
3.2 Materials and Method ........................................................................................... 18
3.2.1 Logistic Regression ........................................................................................ 21
3.2.2 Continuous linear plateau-model ................................................................... 22
3.2.3 Weight category analysis ............................................................................... 22
3.3 Results ................................................................................................................... 23
3.3.1 Descriptive statistics ...................................................................................... 23
3.3.2 Breeding Company 1...................................................................................... 24
3.3.3 Breeding Company 2...................................................................................... 29
3.4 Discussion ............................................................................................................. 34
3.5 Conclusions ........................................................................................................... 37
ix
Chapter 4: A high nutrient specification diet at 9 weeks of age does not improve the
performance of low birth weight pigs ............................................................................. 38
4.1 Introduction ........................................................................................................... 38
4.2 Materials and method ............................................................................................ 39
4.2.1 Experimental design ....................................................................................... 39
4.2.2 Farrowing, lactation and weaner management............................................... 39
4.2.3 Experimental management ............................................................................. 40
4.2.4 Statistical analysis .......................................................................................... 43
4.3 Results ................................................................................................................... 43
4.3.1 Performance in period 1, d 1 to 49 ................................................................. 43
4.3.2 Performance in period 2, d 49 to 63 ............................................................... 44
4.3.3 Performance in period 3, d 63 to 91 ............................................................... 47
4.4 Discussion ............................................................................................................. 50
4.5 Conclusions ........................................................................................................... 53
Chapter 5: Management strategies to improve the performance of low birth weight pigs
to weaning and their long term consequences................................................................. 54
5.1 Introduction ........................................................................................................... 54
5.2 Materials and Method ........................................................................................... 55
5.2.1 Experimental design ....................................................................................... 55
5.2.2 Animal management ...................................................................................... 55
5.2.3 Experimental procedures ................................................................................ 56
5.2.4 Behavioural observations ............................................................................... 58
5.2.5 Statistical analysis .......................................................................................... 58
5.3 Results ................................................................................................................... 60
5.3.1 Performance and behaviour of low birth weight piglets (LBiW) in L or MX
litters ........................................................................................................................ 60
5.3.2 Performance of low (LBiW) and normal birth weight (NBiW) pigs in mixed
litters ........................................................................................................................ 64
5.4 Discussion ............................................................................................................. 66
5.5 Conclusions ........................................................................................................... 70
x
Chapter 6: High specification starter diets improve the performance of low birth weight
pigs to 10 weeks of age ................................................................................................... 71
6.1 Introduction ........................................................................................................... 71
6.2 Materials and method ............................................................................................ 72
6.2.1 Experimental design ....................................................................................... 72
6.2.2 Animal management ...................................................................................... 72
6.2.3 Experimental management ............................................................................. 73
6.2.4 Statistical analysis .......................................................................................... 77
6.3 Results ................................................................................................................... 77
6.3.1 Performance of low birth weight pigs ............................................................ 78
6.3.2 Comparison of low birth weight and normal birth weight pigs ..................... 81
6.3.3 Economic analysis .......................................................................................... 83
6.4 Discussion ............................................................................................................. 85
6.5 Conclusions ........................................................................................................... 88
Chapter 7. General Discussion ........................................................................................ 89
7.1 Can low birth weight pigs grow as fast as normal birth weight pigs? .................. 90
7.2 Improving the growth performance of low birth weight pigs ............................... 91
7.2.1 The importance of timing ............................................................................... 92
7.2.2 The feed intake and efficiency of low birth weight pigs ................................ 93
7.3 What are the long term effects of interventions on growth? ................................. 95
7.4 Management of low birth weight pigs on farm ..................................................... 96
7.4.1 Preweaning management ............................................................................... 97
7.4.2 Postweaning nutrition .................................................................................... 98
7.4.3 The economic implications of low birth weight pigs ..................................... 98
7.5 Scope for future research ...................................................................................... 99
7.6 Final conclusion and findings from the thesis..................................................... 100
References ..................................................................................................................... 101
xi
List of tables
Table 3.1 Descriptive statistics (mean, SD and total number of pigs) of the datasets used
for risk factor analyses associated with poor growth performance in pigs ..................... 19
Table 3.2 Mean body weight (and SD) for birth, weaning, intermediate and final body
weight categories ............................................................................................................. 25
Table 3.3 Mean and range of absolute growth rate (g/day) of pigs derived from 2
datasets during different lifetime stages .......................................................................... 26
Table 4.1 Diet composition and chemical analysis for the dietary treatments, SP and HP,
offered from d 63 to 91 of age. ....................................................................................... 42
Table 4.2 The effect of body weight categories on the performance of pigs from birth to
d 91. NR were normal birth weight pigs but fed restrictedly (between d 49 to 63), N
were normal birth weight pigs and L was low birth weight pigs. ................................... 45
Table 4.3 The effect of body weight category and diet specification (standard (SP) or
high (HP) amino acid: energy ratio) on the performance of pigs from d 63 to 91.......... 49
Table 5.1 The reasons and numbers removed from the experiment and the number of
antibiotic treatment for scours of low birth weight (LBiW) and normal birth weight
(NBiW) pigs on the experiment. ..................................................................................... 62
Table 5.2 The effect of littermate weight and milk supplementation on the performance
of low birth weight pigs (LBiW) pigs from d 1 to 143 ................................................... 63
xii
Table 5.3 The effect of birth weight and milk supplementation on the performance of
pigs in MX litters (litters with both low birth weight (LBiW) and normal birth weight
pigs (NBiW)) from d 1 to 143 ......................................................................................... 65
Table 6.1 Ingredient composition on an as-fed basis and chemical analysis of the four
diets used ......................................................................................................................... 75
Table 6.2 The effect of starter regime and additional allowance on the growth
performance of low birth weight pigs, from d 1 to d 70 ................................................. 80
Table 6.3 The effect of starter regime on the performance of low birth weight and
normal birth weight pigs at weaning, from d 1 to d 70 ................................................... 82
Table 6.4 The effect of birth weight, starter regime and additional allowance on the
mean cost of feeding a pig from d 28 to 70 and margin over feed ................................. 84
xiii
List of figures
Figure 3.1 Odds ratio and confidence limits for the association of low absolute growth
rate of pigs with month of birth, for dataset 1 from birth weight to final body weight
(155 d of age) interval ..................................................................................................... 27
Figure 3.2 Percentage of pigs that remain, increase or decrease at least 1 body weight
(BW) category from birth weight (BiW) to final BW (FW) at 155 d of age for dataset 1
......................................................................................................................................... 29
Figure 3.3 Odds ratio and confidence limits for the association of low absolute growth
rate in pigs with birth weight (BiW), for dataset 2, from BiW to final body weight (BW)
interval (140 d) ................................................................................................................ 30
Figure 3.4 Odds ratio and confidence limits for the association of low absolute growth
rate in pigs with weaning weight (WW), for dataset 2, from birth weight to final body
(BW) interval (140 d) ...................................................................................................... 31
Figure 3.5 Percentage of pigs that remain, increase or decrease at least 1 body weight
(BW) category from birth weight (BiW) to final weight (FW) at 140 d of age for dataset
2 ....................................................................................................................................... 33
Figure 4.1 The effect of body weight (BW) category (N, NR and L), on the scaled ADG
(SDG) of pigs from d 28 to 91 ........................................................................................ 46
Figure 4.2 The effect of body weight category (N, NR and L), on the scaled feed intake
(SFI) of pigs from d 28 to 91. ......................................................................................... 47
1
Chapter 1. Introduction
1.1 What are light pigs?
Within an animal population there will always be natural variation in growth
performance, and pigs are no exception to this. Differences in growth inevitably give
rise to variation in body weight (BW), with the poor performers or light weight pigs
presenting producers with a problem which has significant financial, welfare and
environmental implications (Patience et al., 2004). Whilst the economic impact of light
pigs has resulted in renewed interest in improving their performance, consumer demand
for leaner more consistent meat products has also led to retailers requiring uniform
carcasses. Producers are given tight contract specifications to adhere to with regards to
BW at the abattoir and failure to adhere to these contracts will result in financial
penalties (McCutcheon, 2002). This pressure to produce carcasses within a narrow
weight range means that the industry is very keen to look at ways of increasing the
uniformity of all pigs within a group. Rather than slowing the growth of larger pigs,
increasing the uniformity by increasing the growth of light pigs through, for example,
exploitation of their ability to compensate for previous growth retardation (Handel and
Stickland, 1988) is a preferable approach.
Light pigs can be defined as pigs that, at a certain age, have a BW which is significantly
below the average of the group. They grow markedly slower than their counterparts,
despite the presence of freely available, good quality food that seems to support the fast
growth of their counterparts. This reduction of growth can occur at any stage of
production from birth through to finishing. Whilst this variability in postnatal growth
may arise from management deficiencies, it can also be the result of pigs born with low
birth weight (LBiW) (Quiniou et al., 2002). Often these initial deficits in BW are
exacerbated by the postnatal environment, with access to inferior teats, competition
from heavier littermates and unsuitable nutrition, meaning these pigs are simply unable
to catch up. These LBiW pigs have a significant number of implications for the pig
industry, especially because they may be associated with poor feed conversion
efficiency (FCE) (Powell and Aberle, 1980; Gondret et al., 2006) and increased days to
slaughter (Beaulieu et al., 2010). In recent years selection for sow prolificacy has
resulted in increased litter size, which ultimately causes a decrease in the overall mean
piglet birth weight (BiW) and increased within-litter variability (Roehe, 1999; Quiniou
2
et al., 2002). Consequently, it is likely that as the number of LBiW pigs continues to
increase, the problems faced by producers will worsen. Despite being recognised by the
pig industry as a significant problem, there is a scarcity of literature which investigates
the lifetime performance of light pigs, rather focusing on their growth over shorter
periods such as birth to weaning. It is therefore important to identify the risk factors
associated with poor lifetime performance in pigs, and determine which factors, BiW or
otherwise, contribute significantly to light pigs at slaughter age.
1.2 Problems associated with light pigs
Management of a population of pigs with large differences in BW can be problematic
and result in system inefficiencies. Often, at different stages of production, pigs will be
sorted by BW with the aim of managing the variation. This inevitably requires both staff
and possible sorting equipment (such as automatic sorters) and profitability can be
impacted by sorting and regrouping pigs because of the resultant effect of increased
aggression on performance. Light pigs increase the variability within a group and this
can be associated with inefficient pen utilisation in batch systems and/or financial
penalties at the abattoir for poor grading specification. Farmers are required to produce
pigs within an ideal weight range, with financial penalties at the abattoir if pigs fall
above or below this (Brumm et al., 2002). In a continuous flow system light pigs can be
held back until they reach market weight, but this results in losses by the producer due
to inefficient pen utilisation and additional feed costs. In contrast, in an all in all out
system, light pigs may dictate when the whole shed can be emptied, resulting in a
slower barn throughput. However, holding the main population of pigs too long may
result in the larger pigs exceeding contract limits and having deteriorating FCE.
Light pigs not only pose management difficulties and potential financial losses for
producers, but they also represent a welfare issue for the animals. The highest risk of
mortality for pigs is during the preweaning period (Roehe and Kalm, 2000), with piglet
BiW an important factor for survival (Kerr and Cameron, 1995; Roehe and Kalm,
2000). Wu et al (2006) have reported that the likelihood of survival decreases from 95
to 15% as BiW decreases from 1.80 to 0.61 kg, meaning that light pigs require
additional attention from producers. Whilst there is limited research on the effect of low
BW on post weaning mortality, Larriestra et al (2006) found pigs weighing less than 3.6
kg at weaning (at d 17) had a significantly higher risk of mortality during the nursery
3
period than heavier pigs. However the effects may vary in pigs which have a higher age
at weaning.
For this reason, as well as possible effects on lifetime performance, some producers
may eliminate light pigs from production at birth (Rehfeldt and Kuhn, 2006); however
public perception of this practice is negative. Whilst preweaning and early post weaning
mortality is increased in light pigs, the greatest effect is only seen in those with
extremely low BW, and it is therefore possible that many of these disadvantaged pigs
can benefit from additional care. Not only does this raise welfare questions about the
necessity of euthanasia of a larger number of piglets, but also indicates unnecessary loss
of profits for producers.
1.3 How has the industry attempted to deal with light pigs?
Whilst it is now well established that light pigs are problematic, attempts to deal with
them thus far have had a limited effect. Most commonly, on-farm techniques are
employed to attempt to manage the variation rather than reduce it. As previously
discussed, in more extreme cases, culling piglets under a certain weight at birth has been
tried, but with limited success and negative public perception. Most commonly, the
industry attempts to deal with such animals by regrouping pigs on the basis of live
weight. Regrouping based on live weight can occur at all stages of production, from
cross-fostering in the farrowing house right through to mixing pens in the finisher
house. The current approach does not seem to confer any benefits on the growth of the
light pigs and for this reason the practice can be perceived as disruptive by farmers.
Whilst this process may remove some of the barriers that are imposed on the
performance of lighter pigs, such as issues of competition, sorting by weight appears to
have limited to no beneficial effect (Gonyou and Peterson, 1998; O'Quinn et al., 2001);
despite this it is still common practice on many UK farms. The absence of benefits from
the regrouping of pigs according to their BW may be due the fact that, whilst
regrouping based on live weight may remove competition from heavier littermates, light
pigs are still fed the same feed and kept in the same environmental conditions which
may not be optimal for their reduced BW.
More recently, nutritional interventions at different stages of production have been the
focus of attempts to improve the performance of light pigs. Weaning weight (WW) has
4
a significant effect on subsequent growth performance, with lighter pigs at weaning
performing less well (Kavanagh et al., 1997; Mahan et al., 1998). Manipulation of
preweaning nutrition is therefore one possible approach to improve WW. Wolter et al
(2002) reported that the provision of supplementary milk improved WW; however since
BiW had a greater impact on post weaning performance it was not an effective strategy
to improve the long term performance of light pigs. In a similar study by Morise et al
(2011), provision of a high protein milk replacer to LBiW pigs benefitted their
preweaning growth although the effect only persisted in males post weaning. This
suggests that providing additional nutrition preweaning may not always result in long
term benefits for light pigs or in reduced overall weight variation.
In contrast, the use of specialised starter regimes to improve the nursery performance of
light pigs at weaning has provided more positive results. Most recently, Beaulieu et al
(2012) reported that LBiW pigs responded positively to a complex diet, incorporating a
range of more digestible ingredients rather than a simple diet formulation, with an
improvement in the immediate post weaning performance; however the effects were
only observed for 1 week (wk) post treatment. A study by Magowan et al (2011) which
followed pigs to finishing found that pigs with light WW benefitted from a high
allowance of starter diets, which gave an improved 15 wk weight. This is similar to
Mahan et al (1998) who reported that feeding a phase 1 diet for a longer period was
beneficial, however it was found that WW had a greater influence on post weaning
growth than any dietary regime. So whilst it seems apparent that light pigs may benefit
from specialised nutrition, often these benefits are not present in the longer term or of
great enough magnitude to enable pigs to catch up with the weight for age of heavier
littermates; however this remains an area that requires more research.
As discussed, although light pigs are recognised by the pig industry as a significant
problem, attempts to improve their performance have thus far had a limited effect.
Identifying and understanding the factors which contribute to poor performance and
light pigs is vital in developing treatments to reduce weight variation. It is also
important to establish at what age, or stage of production, light pigs can benefit from
interventions, as it appears that treatments given at some stages are more effective than
at others.
5
1.4 Thesis aims
The aim of this thesis was to provide understanding of risk factors that are associated
with the occurrence of light weight pigs and, by providing such understanding, to
develop nutritional and management treatments that might enable these pigs to decrease
the deficit in their body weight.
The specific objectives of this thesis were:
To identify the risk factors associated with occurrence of light pigs through a detailed
literature review; this will include both pre and postnatal factors which may affect
postnatal performance (Chapter 2).
To conduct a detailed epidemiological study to identify the risk factors which contribute
towards poor performance in pigs and determine whether light pigs can exhibit catch up
growth and at which stages of production (Chapter 3).
To determine if the provision of a high nutrient specification diet will improve the
growth performance of pigs which are light at 9 weeks of age as a result of different
causal factors (chapter 4).
To determine whether management and nutritional treatments preweaning can increase
the WW and performance to slaughter of LBiW pigs (Chapter 5).
To investigate whether a high specification starter regime and the provision of an extra
amount of feed can improve the nursery exit weight of pigs which were light at weaning
as a result of LBiW (Chapter 6).
6
Chapter 2. The risk factors associated with poor growth performance
in pigs
An understanding of the risk factors for poor growth which are inherent in the pig or
present in its environment is critical in reducing variability. Whilst some factors will be
hereditary, many will be the result of the animals’ interactions with the pre- and
postnatal environment. It is important to consider that it is rarely one factor acting
independently which results in suboptimal growth, but several acting simultaneously in
an additive or interactive way (Black et al., 2001).
2.1 Animal characteristics
The homogeneity of growth of pigs can be influenced by genetics, with average daily
live weight gain (ADG) being a moderately heritable trait in pigs at approximately 20 to
40% (van Wijk et al., 2005; Hoque et al., 2007; Rothschild et al., 2011). As the main
aim of pig production is to produce quality lean meat in the most efficient way, the
objectives of breeding programmes commonly include improved leanness, ADG and
FCE (Rothschild et al., 2011). The UK relies on relatively few breeds in commercial
production, with the Large White, Landrace, Duroc and Meishan breeds usually being
cross-bred to produce dam lines with desirable traits. These are then crossed with
purebred or synthetic sire lines to produce the slaughter generation. This crossing of
different breeds, favoured for different production traits, can generate variation in size
and performance in the crossbred slaughter population.
Furthermore in pigs, as well as other animals, males tend to grow faster than females
and are usually heavier when mature (Comstock et al., 1944), although this is not
necessarily supported by literature with conflicting evidence on the effect gender has on
the performance of modern commercial pigs during different stages of production. In
the majority of cases, the effect of sex on performance is not the sole cause of
investigation, but rather how sex interacts with other factors. Looking specifically at pre
weaning growth, Skorjanc et al (2007) investigated the effect of both BiW and sex, with
no effect of sex being reported. Dunshea et al (2003) reported very few effects of sex on
the lifetime growth of pigs, with the exception of the initial period post weaning where
light weight gilts outperformed their male counterparts (this applied only to light weight
animals) (Dunshea et al., 2002). In the nursery period, Hill et al (2007) found no effect
7
of sex (gilts and barrows). However focusing on pigs in the grower and finisher,
O’Connell (2006) found that boars grew both faster and more efficiently than gilts from
60 kg to 100 kg, with an even greater difference in performance noted when dietary
lysine concentrations were increased. Whilst in the UK it is not common practice to
castrate pigs like elsewhere in Europe, as there may be differences in growth between
boars and barrows (Quiniou et al., 2010) this is important to consider when comparing
literature. Focusing on the later stages of production, Quiniou et al (2010) found no
difference in the growth performance of boars, gilts or barrows post weaning from day
(d) 28 to 63; however in the later part of the finisher period from d 105 to 152, gilts and
barrows had significantly lower ADG than boars as well as a poorer FCE. In contrast,
Wolter and Ellis (2001) reported differences in the performance of gilts and barrows in
the finisher stage, with barrows requiring fewer days to reach market weight (Wolter
and Ellis, 2001). Given the lack of clarity, it is important to establish the role sex has on
the effect of lifetime growth performance.
2.2 The prenatal environment
As a polytocous species, the sow uterus must support the growth of a large number of
embryos, development of which requires delivery of vital nutrients and oxygen from the
dam via the placental vascularisation. It is well established that uterine capacity and
insufficient vascularisation (Wu et al., 2008; Oksbjerg et al., 2013; Pardo et al., 2013a)
can become limiting factors for embryo growth and development and result in
intrauterine growth restriction (IUGR) in pigs. IUGR is defined as impaired growth of
the mammalian embryo/foetus, with pigs exhibiting more severe naturally occurring
IUGR than any other domestic species (Wang et al., 2008). Historically, IUGR pigs
have been identified on the basis of their BiW (Bauer et al., 1998), for example 1.5 to
2.0 standard deviation (SD) units below the average BiW (Wang et al., 2009; D'Inca et
al., 2010a; D'Inca et al., 2010b), however it is important to realise that not every LBiW
pig has experienced IUGR; they may be constitutionally small as a result of genetics.
This is an important distinction to make, as IUGR is associated with a greater range of
morbidities and mortalities than occur in piglets which are just small at birth (Baxter et
al., 2008). Therefore it is important to consider both IUGR and BiW separately, as it is
possible that they have differential effects on postnatal piglet performance. However, as
the majority of literature uses the two terms interchangeably, it is often difficult to
disentangle which morbidities are associated with either condition. For this reason, the
8
ponderal index (PI) is often used as an indicator of IUGR. Ponderal index is the ratio of
BW to length (weight/length3), with a low PI (disproportionally long and thin) possibly
indicating IUGR (Chellani et al., 1990).
2.2.1 Intrauterine growth restriction
In the case of unaffected offspring with ‘normal’ prenatal growth, intrauterine space
restriction is minimal which allows normal placental development and consequently an
improved nutrient exchange between foetus and the dam (Pardo et al., 2013a). The
severity of restriction in utero is likely to vary between pigs and therefore not all are
similarly affected, giving rise to varying degrees of IUGR. More recently head
morphology (Chevaux et al., 2010) rather than BiW has been used to identify IUGR. In
particular the ratio of the BW to head size (Baxter et al., 2008) has been used to indicate
a ‘dolphin-like’ head shape (Hales et al., 2013), which is due to the ‘brain sparing’
effect (Bauer et al., 2003) as the body directs nutrients preferentially to the brain to
ensure development of important organs.
At birth the proteomes of the small intestine, liver and skeletal muscles are altered in
IUGR neonates, which may contribute to reductions in immune function, protein
synthesis and cellular signalling (Wang et al., 2008). Pardo et al (2013) also found that
IUGR can affect both the development of internal organs as well as myogenesis.
Inevitably, as a result of altered physiology, IUGR pigs may not respond as well to the
postnatal environment as non-affected pigs. For example, it has been demonstrated that
new born piglets with IUGR have been shown to ingest insufficient amounts of
colostrum compared to normal littermates (Amdi et al., 2013), which may be one
possible cause of increased mortality associated with IUGR pigs.
Prevention of IUGR is a topic with increasing popularity. As most gestating sows are
fed restrictedly, it has been hypothesized by numerous authors that increasing feed
allowance may rectify the problem by reducing nutrient restriction of embryos. Dwyer
et al (1994) found no difference in the BiW of piglets whose mothers had been fed
different diets during early gestation (ranging from d 25 to 80), which is not surprising
as Noblet et al (1985) showed no effect of maternal nutrition on foetal BiW before d 80
of gestation. Although Dwyer et al (1994) found that increasing sows feed intake (FI) at
targeted points in gestation can increase the ration of secondary to primary muscle
9
fibres in piglets (but no increase in total muscle fibre number), more recent work has
found no such effect (Nissen et al., 2003; Bee, 2004), with no benefits observed in
postnatal performance of piglets (Cerisuelo et al., 2009; McNamara et al., 2011;
Rehfeldt et al., 2011). Given the lack of effect of maternal nutrition during gestation,
current thinking indicates that causal factors occur earlier, even prior to ovulation
(Foxcroft et al., 2006). Increasing the period in between weaning and the next
pregnancy in sows has been shown to reduce the coefficient of variation (CV) and
increase the total litter weight at birth, likely related to follicular development (Wientjes
et al., 2013). However at present there are no strategies to successfully prevent IUGR,
especially in more prolific sow lines, so ultimately management of these pigs is
required. Despite all of the deficiencies associated with IUGR, it has been suggested
that pigs may be able to compensate if given special attention and a postnatal
environment adapted to their altered requirements (Wu et al., 2006).
2.2.2 Birth weight
The majority of swine literature suggests an established relationship between BiW and
postnatal growth, with piglets born underweight more likely to underperform
throughout life (Rehfeldt and Kuhn, 2006; Rehfeldt et al., 2008; Beaulieu et al., 2010).
Consequently, BiW has long been considered an important trait in pig production
(Rehfeldt and Kuhn, 2006). Whilst it is inevitable that a number of LBiW pigs in studies
will have been exposed to a certain degree of IUGR, when comparing them to pigs with
classic IUGR symptoms (such as the ‘dolphin head’), it seems that that they do not
exhibit such a wide range of morbidities and mortalities. However, they do have poor
postnatal growth rates when compared to heavier littermates and it has been suggested
that they may remain stunted throughout their life (Gondret et al., 2005; Rehfeldt and
Kuhn, 2006; Wang et al., 2008). For this reason, piglets with significantly LBiW may
be excluded from rearing by producers (Rehfeldt and Kuhn, 2006), especially as
survival rates decrease from 95% to 15% as BiW decreases from 1.80 to 0.61 kg (Wu et
al., 2006). The lower growth potential has been attributed to the fact that pigs born with
LBiW may have a lower number of muscle fibres formed prenatally (Nissen et al.,
2004; Gondret et al., 2005; Gondret et al., 2006; Rehfeldt and Kuhn, 2006; Paredes et
al., 2013). As the number of fibres is set at birth, and subsequent growth is only by
hypertrophy, this may limit growth performance. Low birth weight has also been shown
10
to result in retardation of the digestive tract both at birth (Wang et al., 2005) and
weaning (Michiels et al., 2013), which is likely to affect how they respond to nutrition.
The perceived impact of BiW on postnatal performance is variable in literature, with
some papers suggesting a negative influence and others less so. It is possible that the
variability in the results is due to the definition of LBiW pigs, which can vary between
studies, with more extreme effects on performance only observed in the lowest weight
pigs. Identification of subpopulations which may have the capacity for higher
subsequent growth, versus pigs that are likely to remain stunted, is therefore important.
2.3 The postnatal environment
2.3.1 Lactation
The initial period following parturition, when the piglet is suckling from the sow, is not
only critical for piglet survival but also for subsequent performance because WW is
likely to influence ADG in later stages (Klindt, 2003). It has been reported by Tokach
(1992) that each additional 0.5 kg at weaning corresponds to an additional 2 kg in BW
by slaughter, emphasising the importance of maximising preweaning growth.
In the first few days following parturition, the new born piglet is extremely vulnerable.
Colostrum intake after parturition is considered one of the major determinants of piglet
survival (Devillers et al., 2011), with insufficient intake a major cause of neonatal
mortality (Edwards, 2002). Whilst colostrum allows the transfer of maternal antibodies
to the piglets, it also contains vital growth factors which promote development of the
gut (Wang and Xu, 1996; Dunshea and Van Barneveld, 2003). It has recently been
shown that IUGR piglets consume an insufficient amount of colostrum (Amdi et al.,
2013), which may therefore hinder gut development and could affect digestion and
absorption of food. Once the available colostrum has been consumed, piglets rely on
sows’ milk for their sole source of energy and nutrients. Weaning weight is closely
related to consumption of sow’s milk during lactation (Lewis et al., 1978) and therefore
high intake is vital to maximise preweaning growth. Low birth weight pigs have been
shown to consume less milk per suckling than heavier littermates (Campbell and
Dunkin, 1982). Since the suckling frequency is fixed for the litter as a whole, this may
affect the total amount they consume. As sows’ milk production capacity is limited,
inevitably there will be competition between littermates. During lactation, sibling
11
competition has a major effect on survival and growth; this includes primary
competition, where piglets fight to establish teat ownership, as well as indirect
competition where heavier pigs may be able to better stimulate their teats to access a
disproportionate share of available nutrients (Milligan et al., 2001b). A fixed teat order
is established soon after birth, with the larger or more dominant piglets usually
accessing the anterior teats (Skok et al., 2007). These teats are the more productive, so
that pigs suckling from the posterior teats would consume less milk and subsequently
exhibit lower daily weight gains (Skok et al., 2007).
Cross fostering is a strategy which is commonly employed on UK farms, with the aim
of manipulating litter size, either reducing or increasing the number of piglets usually to
approximately 10 to 12 piglets. Litter size has a significant effect on piglet performance,
with smaller litter sizes having a positive effect on WW (Stewart and Diekman, 1989;
Deen and Bilkei, 2004; English and Bilkei, 2004). This improvement in weight is likely
due to decreased competition for access to a functional teat as there are fewer pigs
(Bilkei and Biro, 1999; Tuchscherer et al., 2000). This is supported by English and
Bilkei (2004) who noted that piglet behaviour varies in larger litters, with smaller pigs
more likely to miss nursing episodes and spend more time in teat disputes.
Cross fostering can also be undertaken to create homogenous litters composed of piglets
with similar BW. This is done on the assumption that LBiW pigs are at a competitive
disadvantage when raised with heavier littermates. Therefore they would be expected to
perform better in litters with smaller pigs and less weight variability, as there is less
competition for access to the best teats, as the biggest pigs are likely to suckle from the
most productive teats (Fraser and Jones, 1975). In support of this, English and Bilkei
(2004) reported a decrease in the 21 d WW of LBiW pigs when grouped with heavier
littermates. However, a series of studies by Milligan et al (Milligan et al., 2001a;
Milligan et al., 2001b; Milligan et al., 2002a; Milligan et al., 2002b) found no difference
in the ADG of LBiW piglets when grouped with heavier pigs, questioning whether they
are indeed disadvantaged. This was supported by Fix et al (2010), who reported no
difference in ADG resulting from within-litter variation. It is important to consider that
while creating homogenous litters will benefit small pigs, which are out competed for
the most productive anterior teats, there will still be a hierarchy in litters composed of
all LBiW or all normal birth weight (NBiW) pigs.
12
2.3.2 Weaning
Natural weaning is a process that occurs gradually, with pigs becoming less reliant on
sow milk and more reliant on other food sources; this usually occurs from 8 to 9 weeks
post-partum and can last for 3 to 4 weeks (Whittemore and Kyriazakis, 2006). However
in today’s commercial farms, weaning is an abrupt and often stressful process which
usually occurs between 21 and 28 d of age, this coincides with milk production reaching
a plateau in the sow at approximately 21 d. Weaning may involve a significant number
of changes to the pigs environment, including change in feed (liquid to solid), transport,
mixing with unfamiliar pigs and new housing (Berkeveld et al., 2007). Many will
experience a post weaning ‘growth check’, where they exhibit loss of weight or a
reduction in weight gain, due to a reduction in FI (Cooper et al., 2009). This is a
widespread problem experienced on farms and results in economic losses for the
producer. This growth check can lead to delays in reaching market weight (Wiseman,
1998), so to ensure minimal growth loss there needs to be high FI of a nutrient dense
diet almost immediately following weaning (Lawlor et al., 2002). Post weaning FI is
likely to be influenced by a number of factors such as the use of creep feed, stress,
temperature, water availability, as well as the nature of feed itself. Ideally starter
regimes need to be highly palatable and digestible, as well as dense in nutrients to
account for the low FI by newly weaned pigs. Both increased allowance and more
complex starter diets including a range of highly digestible ingredients can result in
improved performance (Mahan et al., 1998; Mahan et al., 2004; Magowan et al., 2011b)
as well as a reduction in the numbers of days to slaughter (Mahan et al., 1998).
There has been a gradual decrease in the age at which pigs are weaned over the years
mainly facilitated by the drive to improve sow output through greater farrowing
frequency. More recently, a further driver to reduce weaning age has been an approach
to improve the health of piglets by reducing the transfer of growth depressing pathogens
from sow to piglet in segregated early weaning (SEW) systems (Main et al., 2004).
Whilst individual producers will make a decision about what age is best to wean their
piglets based on sow performance, herd health, pig performance as well as costs and
other factors (Smith et al., 2006), significantly reducing the age may have a negative
effect on both short term and long term pig performance (Edwards, 2010). This is
supported by the results of Leibbrandt et al (1975) who demonstrated that increased
weaning age from two to four weeks resulted in greater adaptation to the post weaning
environment and a decrease in growth check. When considering long term effects, Main
13
et al (2004) found that every one day increase in weaning age from d 12 to 21.5 resulted
in a 1.8 kg increase in weight per pig sold, with differences in performance noted not
only in the initial period after weaning but also long term. It is therefore important for
producers to consider the increased benefits that arise from weaning at an earlier age in
comparison to the negative effects on piglet lifetime performance that it may cause.
2.3.3 Post weaning
Inevitably management strategies vary between farms and the producer is likely to try
and maintain an environment which is suitable for their livestock. However, there are a
number of known environmental factors that can hinder growth performance and, while
the extent of each will vary between holdings, the basic principles remain.
Nutrition is one of the most important factors in growth performance and, as feed
accounts for approximately 60% of the total cost of production (BPEX, 2013b), it is
important that producers get it right. Poor feeding regimes such as incorrect diets,
dramatic changes in diets and inadequate access to feed can all result in reduced FI or
utilisation and a drop in performance. Provision of a diet which does not meet the
nutritional requirements of pigs is likely to affect all pigs, but inevitably will have a
greater effect on some than others. Pigs with reduced growth rates can also find the
problem exacerbated due to competition from dominant pen mates; the more dominant
(often heavier) pigs will take a higher proportion of the food (Baldwin and Meese,
1979), thereby increasing their own growth at the expense of others which will further
increase within pen variation. Any problems in the diet are more likely to have a
pronounced effect in those younger pigs such as weaners, who are at a critical stage in
their digestive maturity, rather than finisher pigs (Whittemore and Kyriazakis, 2006).
Formulation of dietary regimes needs to consider the stage of growth (or BW), genetic
potential as well as the environment of the pig to prevent insufficient or excessive
feeding of nutrients which can be costly. Diets are formulated for the ‘average’ pig in a
population, meaning pigs which fall considerably below this weight are likely to be
disadvantaged. For example, LBiW pigs may have a reduced FI post weaning (Nissen
and Oksbjerg, 2011) and therefore when given the same feed as heavier pigs will
consume less and as a result consume fewer nutrients. These pigs may also have
immature digestive systems, including reduced secretion of digestive enzymes (Xu et
al., 1994; Wang et al., 2008; D'Inca et al., 2010b) and a less developed digestive tract
14
(Michiels et al., 2013) which can hinder digestion and absorption of nutrients. Low birth
weight pigs may also have altered body composition in relation to heavier littermates
both at birth and slaughter, with a higher percentage of adipose tissue (Bee, 2004;
Rehfeldt and Kuhn, 2006; Rehfeldt et al., 2008), which may alter their amino acid:
energy requirements. The effect of feeding light weight pigs a feed which is not
formulated for their requirements means they are unlikely to be able to grow to their full
potential and will perpetuate poor ADG.
Pigs will experience their optimum growth performance when in their thermoneutral
zone (Baker, 2004); therefore in indoor pig rearing environmental temperature is tightly
controlled during each stage of production to optimise growth (Straw et al., 2006). Pigs
do not have the ability to sweat sufficiently, so are susceptible to overheating as they
rely on panting to dissipate heat (Dewey et al., 2009). When exposed to higher ambient
temperatures, a decrease in their FI is observed as the pigs try to reduce heat production
(Renaudeau et al., 2008); subsequently this reduction in nutrients will detrimentally
affect the growth of the pig. A decrease in the ambient temperature below the
thermoneutral zone means pigs require more energy to maintain body temperature. This
will result in an increase in their FI, but no increase in their growth as they metabolise
food to generate heat. The BW and FI of the animal will determine their upper and
lower critical temperatures (Baker, 2004), with lower environmental temperatures for
pigs required as they progress from birth to finishing. As smaller pigs will have a larger
surface area: volume ratio (Stanton and Carroll, 1974) as well as a lower FI, they are
more vulnerable than a larger pig to lower temperature because they will have a greater
lower critical temperature. In weaners, the critical period (0 to 2 weeks post weaning) is
a time where a higher ambient temperature is needed to counteract the thermal
challenge experienced by the pigs due to the reduced FI and metabolism (Le Dividich
and Herpin, 1994). Le Dividich et al (1980) also reported a high incidence of post
weaning diarrhoea with fluctuations of ambient temperature at weaning. Following this
period the temperature can be gradually reduced by 2 to 3°C as the pigs adjust to the
changes imposed at weaning.
The stocking density of pig units can have a significant effect on growth performance,
whilst correct manipulation may be beneficial to pig performance. Under stocking in
colder months can give rise to cold stress if housing has poor ambient temperature
control, whilst decreasing stocking density in warmer months can be useful in reducing
15
heat stress (Jones et al., 2011), which is likely to affect smaller pigs as their lower
critical temperature is greater than larger counterparts. Below a critical threshold,
reducing space allowance reduces FI and growth rate (Gonyou et al., 2006). The
reasons for this may include the fact that overstocking can quickly result in problems
with pigs such as aggression (Jones et al., 2011) and unequal levels of FI, with the
submissive (often smaller) pigs at a disadvantage in accessing resource. As a result there
is a legal minimum space allowance in the EU to prevent overstocking, with the
minimum allowance increasing as the BW of the pig increase.
It is commonplace in pig production for unfamiliar pigs to be mixed; this can occur
from birth to finishing. Mixing can take place for a number of reasons such as
preweaning cross fostering, evening up BW within groups or improving pen utilisation.
For example, at finishing heavier pigs in a pen may be sent to slaughter whilst lighter
pigs that are not a suitable weight for slaughter are held back, and subsequently mixed
with other pigs to clear pens. As there is a social hierarchy present from birth, mixing
pigs will disrupt this leading to the need for pigs to establish a new hierarchy (Puppe et
al., 2008). Whilst mixing can be considered a stressor in pigs, the effect on performance
reported in the literature is variable. Whilst no effect of removing and remixing light
weight pigs post weaning on performance to slaughter was reported by Brumm al
(2002), Spooder et al (2000) concluded that mixing pigs in the finisher stage led to
increased aggression and detrimental effects in the immediate post mixing period. The
adverse effect of mixing on the behaviour of pigs was supported by D’Eath et al (2010),
who reported increased aggression in finisher pigs, although this was in pigs already
classified as having above average aggressive temperament. So, while the aim of mixing
pigs at different stages is usually to reduce the variation in BW, the possible effect on
performance needs to be considered.
2.4 Conclusions
At different stages of production there are a number of different factors which can affect
performance. Not all pigs are born with the same growth potential or experience the
same in utero environment. As such, some pigs may be disadvantaged from birth and
early stages of production with subsequent interactions with the environment likely to
have a significant impact on performance. In particular, the lactation period may inhibit
optimal growth as piglets compete for limited resources. The nutrition of pigs is also
16
likely to have a major impact on performance and pigs with lower BW at different
stages of production may have altered requirements.
Whilst swine literature has identified many risk factors which can negatively affect the
growth performance of pigs, there is a scarcity of data which has taken into account the
effect of these risk factors on the lifetime performance of pigs. Such understanding is
necessary to improve light pig performance, as it is likely that whether light pigs will
benefit from intervention treatments will depend on the reasons that have led to their
lighter weight.
17
Chapter 3: Identification of risk factors associated with poor lifetime
growth performance in pigs
3.1 Introduction
Variation in BW is a common problem in the pig industry and has financial, welfare and
environmental implications (Patience et al., 2004). Variability in postnatal growth may
arise from management deficiencies but can also be a result of pigs born small or
growing markedly slower despite suitable environments. Rather than slowing the
growth of larger pigs, increasing the uniformity by increasing the growth of smaller
pigs, through for example exploitation of their ability to compensate (Handel and
Stickland, 1988), is a preferable approach.
Although research has suggested that numerous factors affect lifetime performance (De
Grau et al., 2005; Larriestra et al., 2006), it is important to consider at what point, or
under what circumstances, a deficit in growth can be considered permanent and
intervention ineffective. Birth weight is considered to be a critical indicator of postnatal
performance, with piglets born underweight often remaining stunted throughout their
life (Gondret et al., 2005; Rehfeldt et al., 2008; Fix et al., 2010). Initial deficits are often
exacerbated by access to inferior teats, stress at weaning, and competition from heavier
littermates, meaning these piglets are simply unable to catch up. However literature also
suggests that WW may be a better predictor of future growth and the number of days it
takes to reach slaughter weight (Wolter and Ellis, 2001; Smith et al., 2007).
There is a distinct lack of data available on lifetime performance of low BW pigs. As a
result, research on how to manage weight variation is limited, with few practical
suggestions that have been proven to be effective (O'Quinn et al., 2001; Brumm et al.,
2002). The aim of this paper was to identify risk factors associated with poor lifetime
performance in pigs, paying particular attention to the effect BiW and WW have on
subsequent performance. Understanding such factors would lead to effective
interventions that may reduce BW variability within a group.
18
3.2 Materials and Method
Two production databases were obtained from commercial breeding companies
operating internationally. The datasets for breeding companies 1 and 2 contained
records for approximately 40,000 and 90,000 pigs respectively, and a range of variables
used for analyses (Table 3.1). Databases included only pigs that survived from birth to
finishing.
Breeding company 1 provided data for 10,181 litters produced in the same unit over the
period January 2000 to June 2010. Only a subset of pigs from each litter was followed,
with 69% of those being gilts and the rest intact males. These data were from pigs
produced in the United Kingdom in a conventional health unit (Enzootic Pneumonia
and Porcine Reproductive Respiratory Syndrome positive) and consisted of 3 different
genotypes (described here as 1, 2, and 3). Genotype 1 was a Large White sire line,
genotype 2 was a Large White dam line and genotype 3 was a Landrace dam line.
Pigs were individually weighed within 24 hours (h) of birth (BiW) and before the
majority reached slaughter weight at approximately 100 kg [final live weight (FW)]; for
approximately one-half of the pigs (20,297) an intermediate BW (IW) when an animal
reached approximately 45 kg was also available. Individual BiW were recorded for all
animals born alive. Cross-fostering took place on the unit; however, piglet movement
between sows was not recorded in the database; usually this was within the first 24 h,
with fostering after this kept to an absolute minimum. Pigs were fed 2 commercial creep
feeds, 1 from 5 to 8 kg and the second from 8 to 15 kg. A “rearer” diet was fed from 15
to 30 kg and finally a standard commercial grower diet from 30 kg to slaughter.
19
Table 3.1 Descriptive statistics (mean, SD and total number of pigs) of the datasets
used for risk factor analyses associated with poor growth performance in pigs
Dataset 1 Dataset 2
Variable Total n Mean SD Total n Mean SD
Birth weight, kg1 43,571 1.41 0.29 91,210 1.39 0.33
Weaning weight, kg2 - - - 38,975 6.11 1.38
Intermediate weight, kg3 20,297 48.6 4.82 - - -
Final weight, kg4 20,307 92.5 10.4 5,217 78.0 11.7
Total no. of piglets born per litter5 43,571 12.1 3.03 91,557 11.5 2.80
Born alive piglets per litter6 43,571 11.5 2.90 91,573 10.9 2.73
Parity7 43,571 2.72 1.53 79,393 3.57 2.45
Month of birth, mo8 43,571 6.50 3.44 91,573 6.38 3.03
Gestation length, d9 - - - 79,393 115 1.56
Percentage males10
43,571 31.5 - 91,572 52.0 -
1 Birth weight (kg) of piglets taken within 24 h of birth.
2 Weaning weight of pigs (kg) adjusted to 23 d.
3 Adjusted intermediate BW of pigs taken at 105 d.
4 Slaughter weight of pigs (kg), adjusted to 155 d for dataset 1 and 140 d for dataset 2.
5 Total number of piglets born per litter, including stillborn.
6 Total number of piglets born alive per litter.
7 The number of litters a sow has produced.
8 Month of birth of each litter (starting as January = 1, December = 12)
9 Gestation period of the sow (days)
10 Percentage of pigs in each dataset which are male; the total number is the total
number of animals in the dataset with sex specified
Data from breeding company 2 were from pigs produced in 3 units, using the same
methods and breeds that are present in the United Kingdom and have previously been
described by Kapell et al (2011). The data were generated between January 2005 and
September 2006. Management conditions were standardised across units; units were
coded and included as a factor in the analyses. Pigs were born out of 13,429 litters with
individual BiW available for all pigs; although the time window within which this was
taken was unspecified, the expectation was that this would have been within 24 h from
birth. A second BW was taken at weaning (WW; approximately 21 to 28 d) and a FW
before slaughter at approximately 80 kg. Only a subset of the pigs with BiW had WW
and only a subset of those had FW. A percentage of males were intact (purebred) whilst
20
the rest of the male pigs were castrated (crossbred). Information regarding cross-
fostering was not available. In addition to individual pig BW, a number of variables
were recorded in both of the datasets (Table 3.1). For the majority of pigs these
variables were available: sow parity number, total number of piglets born per litter
(including born alive and stillborn), date of birth, and sex and number of pigs weaned.
For breeding company 2 the length of gestation of the sow was also available.
Data handling and analysis was undertaken using SAS (SAS Inst. Inc., Cary, NC)
and the same methodologies were applied to both datasets. Incorrect or doubtful
observations were removed or corrected when possible: for example if a BW was
greater or less than 4 SD from the mean, then it was removed. In addition to the existing
variables, a number of new ones were calculated. As there was variation in the age at
which BW were taken, adjusted BW for a specific (average) age was used. Adjusted
BW for age was calculated using this formula:
Daily BW gain = (BW 2 – BW 1) / (age BW 2 – age BW 1)
Adjusted BW for age = (daily BW gain x average age at BW 2) + BW 1
Based on the adjusted BW, both adjusted absolute (g.d-1
) and relative (g.day-1
BW-1
)
growth rate were calculated for each pig for each stage of their life (when available).
The relative growth rate is the difference in natural logarithms of the first and last BW
divided by the time between the 2 BW (Winder et al., 1990) and is commonly used in
breeding experiments to account for differences in initial size. Growth rates for each
individual animal were then grouped into 1 of 3 categories using percentiles (33%)
denoting, high, medium, and low growth rates. Pigs were also retrospectively assigned
to a BiW and a IW/WW and a FW group based on percentiles: 12.5% (8 groups), 25%
(4 groups) and 50% (2 groups; above and below the mean). In the analysis only the
12.5% groups were used as these gave us a greater number of categories to compare.
These groups categorised the BW into groups from low to high (1 denoting the lightest,
8 the heaviest). For all analyses the 2 datasets were analysed separately. Potential
variables were normally distributed so were entered into the models without being
transformed.
Three types of analysis were conducted on the datasets. First an ordinal logistic
regression model using categorised growth rates with covariates as categorical or
21
continuous data. This type of generalized linear model allows the prediction of a
dichotomous outcome from multiple variables. Second, a linear plateau regression
analysis was constructed, where both the dependent and independent variables were
used as continuous variables. This model was chosen to allow estimation of a BW
breakpoint, if present in the population of pigs. Finally, the ability of pigs with different
BW at birth and weaning/intermediate BW to reduce the deficit for low BW by the end
of their productive life was estimated, by determining the percentage of pigs that
decreased, increased or remained in the same BW category from BiW/WW/IW to FW.
3.2.1 Logistic Regression
To identify potential risk factors for poor postnatal growth in pigs an ordinal logistic
regression model was constructed using absolute and relative growth rate. Initially a
univariate logistic regression model was used to identify potential significant risk
factors for entry in to the multivariate model. All potential variables were individually
fitted to the model to identify those that were significant; any variables which were not
significant at 5% level were eliminated from inclusion in the multivariate model. After
univariate analyses all variables which were significant were checked for correlation
with other variables to be entered into the model. If variables were found to be highly
correlated (0.70 or above) then the variable which had the greatest effect in the
univariate model was retained and the other excluded from further analysis.
The absolute and relative growth rate of the pigs from BiW to WW/IW, WW/IW to FW
and BiW to FW were used as the dependent variable in the models. The reference
category for these models was category 3 (high growth rate) compared with the
combined effects of categories 1 and 2. A separate multivariate model was run for each
of the 3 stages of growth and included risk factors which were applicable to that stage.
For example the model from birth to weaning included BiW as an indicator, whereas
weaning to finishing included both BiW and WW.
The PROC logistic method (SAS Inst. Inc) was used to run the logistic regression
model. Variables entered into the model (where applicable) were total number of pigs
born per litter (dead or alive), sex (intact male or female for database 1 and intact,
castrated or female for database 2), sow parity, month of birth, BiW and WW/IW (Table
3.1). Breed code was also inputted in to the model for breeding company 1. Independent
22
variables were added into the model by stepwise entry. All variables were entered into
the model as categorical with the exception of the total number of piglets born per litter
and parity. Separate models were run for each BiW and WW/IW interval. The results of
the logistic model gave odds ratios (OR) as well as 95% confidence intervals. All results
refer to the absolute growth rate from BiW to FW unless specified.
3.2.2 Continuous linear plateau-model
To examine the effect of BiW and WW on the lifetime growth rate of pigs, a continuous
linear plateau model was fitted to the data for all pigs using the NLIN procedure of
SAS. The model was adapted from Piegorsch and Bailer (2005) and consisted of a
single sloping line intersecting a plateau at a break point value. This type of regression
allows for an accurate estimate of the break point and additional variables. The linear
plateau model described the growth rate in relation to the BiW (kg) by this pair of
equations:
Y = a + b x (X - Xmax) If X < Xmax
Y = c if X > Xmax,
in which Y is the dependent variable (either absolute or relative growth rate) and X is
the independent variable (BiW/WW/IW). Parameter a is the intercept while parameter b
is the slope of the line up to point Xmax, which occurs at the intersection of the linear
response and the plateau line, and C is the maximum value of Y, also referred to as the
plateau yield. Initial estimates for the variables were calculated using the large-sample
method (Piegorsch and Bailer, 2005). All data were then fitted to the linear-plateau
model using Proc NLIN with the Levenberg-Marquardt fitting algorithm. Residuals
were computed to check the assumptions of normality and linearity and all factors were
plotted before analysis to examine the relationships. Once the breakpoint had been
estimated using the models, linear regression was applied to the data below the
breakpoint to identify the variables that were acting on the growth rates. Variables
entered into the linear models were the same as previously used (Table 3.1).
3.2.3 Weight category analysis
To investigate whether pigs with different BiW, WW and/or IW had the capacity to
compensate for low BW, all pigs at each BW interval were divided using percentiles
23
into 8 categories as previously described. The category of each pig at FW was then
compared with its BiW/WW/IW category to determine whether it remained in the same
BW category, decreased, or increased at least 1 category.
3.3 Results
3.3.1 Descriptive statistics
Table 3.1 shows the average for variables for breeding company 1. The average BiW
was 1.41 (SD = 0.29, range of 0.40 to 2.50) kg, with an average adjusted FW of 92.5
(SD = 10.4, range of 64.8 to 138) kg at 155 d of age and an adjusted IW of 48.6 (SD =
4.82, range of 27.1 to 67.1) kg at 105 d. The majority of pigs were females (69%), with
males having a slightly greater average BiW of 1.47 (SD = 0.30) compared to 1.39 (SD
= 0.29) kg in females (P < 0.001). By IW females were slightly heavier with an average
of 49.6 (SD = 4.84) kg and males weighing 48.0 (SD = 4.73) kg (P = 0.0012). However,
by FW males were heavier once again at 95.6 (SD = 9.96) kg compared to 86.5 (SD =
8.57) kg in females (P < 0.001). The correlations between BiW and IW/FW were low,
+0.22 and +0.21, respectively (P < 0.001), whereas IW and FW had a greater
correlation of +0.48 (P < 0.001). The total number of piglets born per litter ranged from
2 to 36; however, the maximum total number born alive was 22. The parity number of
the sow varied from 1 to 9, parities 2 to 5 had the greatest average BiW, and this pattern
was still evident at FW.
Table 3.1 also shows the average for variables for breeding company 2. The results for
breeding company 2 showed an average BiW of 1.39 (SD = 0.33; range of 0.32 to 2.70)
kg, adjusted WW of 6.11 (SD = 1.38; range of 1.20 to 11.6 kg) at 23 d, and adjusted FW
of 78.0 (SD = 11.7; range 38.4 to 120) kg at 140 d. There were a similar number of
males and females in the dataset (52 versus 48%). Of the 47,621 males in the dataset,
33% were castrated and 67% were intact. On average males had a slightly greater BiW
of 1.52 (SD = 0.33) versus 1.43 (SD = 0.32) kg in females (P < 0.001). By weaning,
males were still heavier than females at 6.16 kg (SD = 1.37) versus 6.06 kg (SD = 1.33;
P < 0.001); however by FW this difference had increased and males had an average FW
of 78.8 (SD = 11.9) compared with females at 77.0 kg (SD = 11.4; P < 0.001). The
correlations between BiW and WW/FW were low, +0.31 and +0.29, respectively (P <
0.001), and WW and FW were also weakly correlated with +0.29 (P < 0.001). The total
number of piglets born per litter ranged from 1 to 25 whereas the maximum total
24
number of piglets born alive per litter was 22. The parity number of the sow varied from
1 to 12, parities 1 to 8 had the greatest average BiW, and this pattern was still evident at
FW.
3.3.2 Breeding Company 1
The variables that were significant in the multivariate logistic model on absolute growth
during the interval BiW to FW (i.e. lifetime growth) were BiW (P < 0.001), IW (P <
0.001), sex (P < 0.001), month of birth (P < 0.001), and breed code (P < 0.001); parity
and total born were not significant. For the additional models looking at BiW to IW and
IW to FW, the same variables were significant, with the addition that parity (P < 0.001)
and total born (P < 0.002) were also significant (OR = 1.15 and 1.02, respectively) for
the BiW to IW interval. Tables 3.2 and 3.3 show the average BW of each BW category
(1 to 8) and the range and average absolute growth rate of each growth rate category (1
to 3) used in the analysis.
Pigs in the lightest BiW category were 1.42 times (P < 0.001) more likely to be in a low
growth rate group than pigs with a BiW of 1.95 kg or above; all other BiW categories
were not significant with the exception of the 2 heaviest, which had even odds (P <
0.001) of being in a low growth rate group compared with the reference category. The
odds of a pig with a lighter IW (31.1 to 43.1 kg) being in a low growth category were
48.9 times (P < 0.001) more likely than a pig with a heavier IW (54.0 to 66.8 kg).
Similarly for the interval IW to FW, pigs with the lightest IW were 50.7 times (P <
0.001) more likely to be in a low growth rate group. Males were less likely than females
to be in a low growth rate category for BiW to FW interval and IW to FW (OR = 0.10
for both intervals; P < 0.001); however between BiW and IW, males were 1.62 times (P
< 0.001) more likely to be in a low growth rate group compared with females. The
results of the breed of pig indicated that the odds of a pig that is classified as breed code
3 being in a low growth rate category were 1.50 times (P < 0.001) more likely than
breed code 1 or 2.
25
Table 3.2 Mean body weight (and SD) for birth, weaning, intermediate and final body
weight categories
BW category1 Dataset BiW
2 SD WW
3 SD IW
4 SD FW
5 SD
1 1 0.93 0.10 - - 40.8 2.07 76.0 3.37
2 0.86 0.12 4.03 0.41 - - 58.0 5.07
2 1 1.20 0.05 - - 44.3 0.64 83.0 1.30
2 1.08 0.05 4.86 0.17 - - 67.6 1.74
3 1 1.30 0.01 - - 46.2 0.48 87.0 1.07
2 1.22 0.03 5.36 0.13 - - 72.6 1.22
4 1 1.40 0.01 - - 47.8 0.42 90.5 0.99
2 1.33 0.03 5.81 0.13 - - 76.6 1.10
5 1 1.60 0.04 - - 49.2 0.43 93.8 0.96
2 1.43 0.03 6.25 0.13 - - 80.1 0.98
6 1 1.70 0.01 - - 50.8 0.48 97.3 1.08
2 1.54 0.03 6.72 0.14 - - 83.5 1.06
7 1 1.90 0.04 - - 52.8 0.69 102 1.59
2 1.68 0.05 7.30 0.20 - - 88.1 1.64
8 1 2.00 0.13 - - 56.7 2.34 110 4.83
2 1.95 0.15 8.47 0.72 - - 97.1 5.10 1
Body weight categories were calculated by retrospectively assigning pigs to a group
based on their weight at each successive BW (i.e. BiW, IW, and FW) using percentiles
of 12.5% (8 groups)
2BiW = birth weight (kg) of piglets taken within 24 h of birth
3 WW = weaning weight of pigs (kg) adjusted to 23 d
4 IW = intermediate weight of pigs adjusted to 105 d
5 FW = final BW; body weight of pigs taken at slaughter (kg), adjusted to 155 d for
dataset 1 and 140 d for dataset 2
26
Table 3.3 Mean and range of absolute growth rate (g/day) of pigs derived from 2
datasets during different lifetime stages
Dataset Time interval Absolute
growth rate
category1
No. of
observations
Absolute
growth rate
range (g/d)
Mean
growth
rate (g/d)
1
Birth to
intermediate
BW2
1 6,765 287-433 402
2 6,761 433-471 452
3 6,771 471-625 504
Intermediate
to final BW3
1 6,765 207-793 665
2 6,766 793-962 880
3 6,766 962-1,607 1088
Birth to final
BW4
1 6,769 413-559 518
2 6,769 559-618 589
3 6,769 618-852 666
2
Birth to
weaning BW5
1 12,970 49-186 152
2 12,974 186-237 212
3 12,974 237-466 279
Weaning to
final BW6
1 1,717 268-575 506
2 1,718 575-655 616
3 1,717 654-987 717
Birth to final
BW7
1 1,717 268-517 458
2 1,718 517-586 553
3 1,717 586-857 640 1
Pigs were categorised into 1 of 3 growth rate categories according to their absolute
growth rate for each individual period of growth. Each of the categories contained a
similar number of pigs and the number of pigs per interval differed depending on pigs
available; for a detailed description see text.
2 Growth rate for this period was calculated using the birth weight (kg) of piglets taken
within 24 h of birth and the intermediate BW of pigs (kg) adjusted to 105 d
3 Growth rate for this period was calculated using the intermediate BW of pigs (kg)
adjusted to 105 d and the BW of pigs at slaughter adjusted to 155 d 4
Growth rate for this period was calculated using the growth from birth to intermediate
BW and then from intermediate to BW 5
Calculated using the birth weight (kg) of piglets taken within 24 h of birth and the
weaning weight of pigs adjusted to 23 d
6
Calculated using the weaning weight of pigs (kg) adjusted to 23 d and the BW of pigs
taken at slaughter (kg) adjusted to 140 d 7
Growth rate for this period was calculated using the growth from birth to weaning
weight and then from weaning weight to final BW
27
Figure 3.1 shows the relationship between the month of birth and subsequent growth
rate. Piglets born during the months of January to April were more likely to be in a low
growth rate group (when compared to December), with the greatest odds of 1.26 (P <
0.001) occurring in March, and piglets born in all other months were more likely to be
in a high growth rate group (when compared to December).
The results for the same models using relative growth rate showed similar patterns for
all variables as previously described, with the exception of BiW. Pigs in the lightest
BiW category were 0.001 times less likely (P < 0.001) to be in a low category for
relative growth rate from birth to finish than the heaviest pigs.
Figure 3.1 Odds ratio and confidence limits for the association of low absolute growth
rate of pigs with month of birth, for dataset 1 from birth weight to final body weight
(155 d of age) interval. The months January to November are compared to the reference
category; this was set as the last month of the year, December. The error bars represent
confidence intervals.
0.00
0.20
0.40
0.60
0.80
1.00
1.20
1.40
1.60
Od
ds
rati
o
Month of birth, m
28
The linear-plateau model showed a breakpoint of 1.91 kg [Confidence Interval (CI) =
1.84 to 1.95] when BiW was plotted against lifetime growth rate (P < 0.001). This
estimation was not affected by the 3 genotypes considered. Linear regression applied
before the breakpoint indicated that IW was the best predictive factor for lifetime
growth rate for those pigs born under 1.91 kg (r² = 0.23; P < 0.001). These variables
were also significant predictors of postnatal growth with decreasing r² value: sex (r2 =
0.19; P < 0.001), month of birth (r2
= 0.03; P < 0.001), and BiW (r2
= 0.02; P > 0.001).
There was no breakpoint for IW vs. lifetime growth rate or vs. IW to FW growth rate.
Figure 3.2 shows the percentage of piglets that remained, increased, or decreased at
least 1 BW category from BiW to FW. Piglets in the lightest BiW category had the
greatest percentage of increases (74%) whereas the pigs in the heaviest BiW category
experienced the greatest percentage of decreases (79%). A similar pattern was observed
for the periods BiW to IW and IW to FW. Of those piglets in the lightest BiW category
that increased at least 1 category from BiW to FW, 47% had reached or exceeded the
BW group (4) by FW.
29
Figure 3.2 Percentage of pigs that remain, increase or decrease at least 1 body weight
(BW) category from birth weight (BiW) to final BW (FW) at 155 d of age for dataset 1.
All pigs were categorized into 8 BiW groups using percentiles resulting in a similar
number of pigs per category; pigs were then categorized at FW using the same method.
A pig is categorized as “remain” when it has not changed BW category from BiW to
FW. A pig is categorized as an “increase” when it has increased at least 1 category from
BiW to FW. A pig is categorized as “decrease” when it has decreased at least 1 category
from BiW to FW. Piglets in the top and bottom BiW categories cannot increase or
decrease, respectively.
3.3.3 Breeding Company 2
The variables that were significant in the multivariate logistic model on the BiW to FW
growth rate were BiW (P < 0.001), WW (P < 0.001), sex (P < 0.001), and month of
birth (P < 0.001); total number born, parity and length of gestation were not significant.
For the additional models looking at BiW to WW and WW to FW growth rates, the
same variables were significant with the addition that for BiW to WW, total number
born (P < 0.001), parity (P < 0.001), and length of gestation (P < 0.001) were also
significant. Table 3.3 shows the range and average absolute growth rate of each growth
rate category (1 to 3) used in the analysis.
0
10
20
30
40
50
60
70
80
90
100
1 2 3 4 5 6 7 8
Per
cen
tage,
%
Birth weight category
Decrease
Remain
Increase
30
The effect of BiW on lifetime growth rate showed that piglets in the lightest BiW (0.32
- 1.00 kg) category were 4.55 times (P < 0.001) more likely to be in a low growth rate
group when compared with those piglets with a BiW between 1.80 and 2.70 kg, with the
odds decreasing as BiW decreased (Fig. 3.3). A similar pattern was observed for all
intervals examined. Figure 3.4 shows the effect of WW on lifetime growth rate; piglets
in the lightest WW category (1.22 to 4.55 kg) were 5.39 times (P < 0.001) more likely
to be in the low growth rate group when compared to the heaviest WW (7.70 to 11.6
kg). Similarly for the interval WW to FW, pigs with the lightest WW were 4.51 times
(P < 0.001) more likely to be in a low growth rate group. Males (intact and castrated)
were consistently less likely to be in the low growth rate group when compared with
females throughout all growth rate intervals examined although during the birth to
weaning interval the odds were almost even (OR = 0.96; P < 0.03). During the interval
WW to FW, castrated males were more likely to be in a low growth rate group
compared with intact males (OR = 1.67; P < 0.002).
Figure 3.3 Odds ratio and confidence limits for the association of low absolute growth
rate in pigs with birth weight (BiW), for dataset 2, from BiW to final body weight (BW)
interval (140 d). Birth weight categories were created by grouping adjusted BiW by
percentiles with a similar number of pigs per category; for a detailed description please
see text. The BiW categories are compared with the reference category; this was set as
the heaviest BiW category which is 1.8 to 2.7 kg. The error bars represent confidence
intervals.
0.00
1.00
2.00
3.00
4.00
5.00
6.00
7.00
Od
ds
rati
o
Birth weight , kg
31
Figure 3.4 Odds ratio and confidence limits for the association of low absolute growth
rate in pigs with weaning weight (WW), for dataset 2, from birth weight to final body
(BW) interval (140 d). Weaning weight categories were created by grouping adjusted
WW by percentiles with a similar number of pigs per category; for a detailed
description please see text. The WW categories are compared to the reference category;
this was set at the heaviest WW category, which is 7.7 to 11.6 kg. The error bars
represent confidence intervals.
The effect of sow parity, although having a significant effect (P < 0.001) on the BiW to
WW growth rate, did not follow a clear pattern for individual parities; this was also true
for piglets born per litter. The length of gestation of the litter was significant (P < 0.001)
on the BiW to WW growth rate, with an increased length of gestation associated with
decreased odds of being in a low growth rate group (OR = 0.98; P < 0.001). The month
that piglets were born did not show any distinct patterns of seasonality for growth rates
with the odds of being in a low growth rate group fluctuating throughout the year.
When using the relative growth rate as the dependent variable in the models, the results
followed a similar pattern as described above with the exception of BiW. Those in the
lightest BiW category were < 0.001 times less likely (P < 0.001) to be in the low growth
rate category than the heaviest BiW category. The odds ratio increased slightly for
successive categories (2 to 7), with the seventh BiW category still 0.16 times (P <
0.001) less likely to be in a low growth rate group than category 8 (1.80 to 2.70 kg). The
0.00
1.00
2.00
3.00
4.00
5.00
6.00
7.00
8.00
Od
ds
rati
o
Weaning weight, kg
32
results for the linear plateau model showed that the breakpoint estimation for BiW
against lifetime growth was 1.84 kg (CI = 1.76 to 1.86) for the BiW of the pigs (P <
0.001). When the linear regression model was fitted to the data before the breakpoint,
WW (r² = 0.07; P < 0.001), BiW (r² = 0.03; P < 0.001), parity (r² = 0.02; P < 0.001),
month of birth (r² = 0.01; P < 0.001), and sex (r² = 0.01; P < 0.001) were significant
predictors of lifetime growth rate. For the model WW versus WW to FW interval, a
breakpoint estimation of 7.52 kg (P < 0.001) was obtained. These variables were
significant predictors of WW to FW and lifetime growth with decreasing r² value: BiW
(r2
= 0.07; P < 0.001), WW (r2
= 0.04; P < 0.001), parity (r2
= 0.18; P < 0.001), month of
birth (r2
= 0.01; P < 0.001), and sex (r2
= 0.01; P < 0.001).
Figure 3.5 shows the percentage of piglets that remained or changed BW categories
from birth to finishing. Whilst the percentage of piglets that remained in the same
category throughout their life was similar for all BiW categories at 10 to 20%, piglets in
the heaviest BiW category had the greatest percentage decrease, meaning that those
piglets with the lighter BiW were more likely to increase BW categories. A similar
pattern was observed for the period BiW to WW; however, for WW to FW a greater
percentage of pigs in the greater WW categories (5 to 8) decreased at least 1 category.
Of those piglets in the lightest BiW category that increased at least 1 category from
BiW to FW, 52% had reached or exceeded the average BW group (4) by FW.
33
Figure 3.5 Percentage of pigs that remain, increase or decrease at least 1 body weight
(BW) category from birth weight (BiW) to final weight (FW) at 140 d of age for dataset
2. All pigs were categorized into 8 BiW groups using percentiles resulting in a similar
number of pigs per category; pigs were then categorized at FW using the same method.
A pig is categorized as “remain” when it has not changed BW category from BiW to
FW. A pig is categorized as an “increase” when it has increased at least 1 category from
BiW to FW. A pig is categorized as “decrease” when it has decreased at least 1 category
from BiW to FW. Piglets in the top and bottom BiW categories cannot increase or
decrease, respectively.
0
10
20
30
40
50
60
70
80
90
100
1 2 3 4 5 6 7 8
Per
cen
tage,
%
Birth weight category
Decrease
Remain
Increase
34
3.4 Discussion
The aim of this paper was to identify risk factors associated with poor growth
performance in pigs and to investigate the relationship between BiW and subsequent
BW to ascertain whether BiW is the more critical factor in determining lifetime growth
rate. In agreement with previous studies (De Grau et al., 2005; Larriestra et al., 2006;
Paredes et al., 2012), the data presented here confirm that poor growth performance is
associated with a number of variables, including LBiW, low WW and IW as well as
litter factors. Additionally, the SD of BiW and FW from both datasets was consistent
with previous literature (Fix et al., 2010; Paredes et al., 2012), confirming the reliability
of our data. In contrast to other studies (Rehfeldt and Kuhn, 2006), lighter BiW piglets
were capable of increasing BW category from birth/weaning to slaughter, suggesting
that these light weight pigs have the potential to reduce the deficit during postnatal
growth. The results were very consistent between the 2 data sets used, allowing some
confidence that the conclusions that can be drawn from them have some generality.
There have been few studies that have investigated more than 2 or 3 factors affecting
lifetime growth performance; the number of observations used in such studies is
relatively small. Our study differs in both these respects. In addition, previous studies
have focused on shorter time intervals, such as the pre-weaning period (Larriestra et al.,
2006), or weaning to slaughter (Wolter and Ellis, 2001), as well as having involved
relatively few variables (Beaulieu et al., 2010). Most such studies usually exclude pigs
considered to have particularly light BiW (usually < 0.8) as they are considered to be
runts and would be expected to influence the value of the conclusions drawn
(Greenwood et al., 2009). By contrast in our study, piglets with BiW as low as 0.3 kg
were considered.
Literature suggests there is an established relationship between BiW and life time
growth rate in the pig, with those born underweight having a greater risk of mortality
(Rehfeldt and Kuhn, 2006), remaining stunted throughout their life (Quiniou et al.,
2002; Gondret et al., 2005; Rehfeldt and Kuhn, 2006) and having poorer meat quality
(Gondret et al., 2006). Although the results of the logistic regression show that those
pigs born with the lightest BiW were more likely to exhibit poor growth to FW, BiW
was not the sole determinant of postnatal growth as both WW, IW and other factors
were significant. This is further supported by the weak correlation between the BiW and
WW/IW, which disagrees with most current literature (Le Dividich, 1999; Quiniou et
35
al., 2002). The exception is the study of Poore and Fowden (2004) , who did not report a
relationship between the BiW of male pigs and subsequent BW at 3 and 12 mo of age.
Additionally, the likelihood of pigs with lighter WW and IW having poor growth to FW
exceeds that of piglets with LBiW, which implies that LBiW has less effect on
performance to finishing in comparison to WW and IW. Furthermore, our results
contradict those of Lynch et al (1998) who suggest that WW was poorly related to post
weaning performance. However, differences in the management of these pigs during the
different stages of production may account for the inconsistent conclusions between
studies (Kyriazakis and Houdijk, 2007).
In addition to BiW and WW, a number of variables which contribute to poor growth in
pigs were identified; these include litter size, sex and month of birth. Although increase
in prolificacy of sows in recent years is seen by many as a positive development, with
increased numbers of piglets being weaned per litter, there are also negative effects
associated with it, the most common being a reduction in the mean BiW of the litter
(Beaulieu et al., 2010). Our results showed that the total number of piglets born per
litter is only significant in determining pig growth during the first interval of their life,
from birth to weaning. During this interval larger litters were more likely to have poor
postnatal growth, similar to previous studies (Beaulieu et al., 2010); however, beyond
weaning the impact of litter size on performance is negligible. Although the effect of
litter size is likely to be partly a reflection of the BiW of the piglets, it can affect their
growth by other routes. Although the sow is able to increase the amount of milk she
produces for a larger litter, this will not be proportional to the number of pigs she
nurses; as a result there may not be sufficient milk for all piglets (Auldist et al., 1998).
Inevitably this can lead to increased competition and mortality, often with the lightest
pigs being most affected by any shortage in milk as they have a competitive
disadvantage to larger litter mates (Milligan et al., 2002b). This is likely to perpetuate
any observed differences in BW during the lactation period ensuring the smallest pigs
stay small.
The parity of the sow can also affect milk production, which is likely to affect piglet
growth. It has been shown that piglets from mid parity sows have the greatest WW
(Milligan et al., 2002a). It can also influence BiW with sows in their first parity having
piglets with lower BiW, as well as fewer piglets (Milligan et al., 2002b). The effect sow
parity has on subsequent piglet growth is unclear. In our study, although parity was
36
identified as a statistically significant factor for growth in all periods of pig life, often
the pattern across the consecutive parities was unclear. The seasonality observed in the
first dataset suggests that the month in which piglets were born in can affect growth
rate. Similar to previous research we found that those pigs born during the warmest
months and therefore finishing during the cooler months were more likely to have
greater growth rates (Kościński et al., 2009). Additionally, sex appeared to be a key
factor for piglet growth, with male pigs exhibiting greater lifetime growth rates in this
analysis, although during the birth to weaning interval the growth rates between the
sexes were very similar (with females actually having higher growth rates in 1 dataset).
It is possible that differences in performance between the sexes are not observed until
post puberty; this is supported by previous work that shows sex did not significantly
affect ADG in piglets during the lactation period (Skorjanc et al., 2007).
It was previously found that the relationship between lifetime growth rate and BiW is
not linear, a threshold can be reached beyond which any further increase in BiW will
not increase absolute growth rate (Fix et al., 2010). Similarly we found that the
relationship between BiW and growth from BiW to FW is linear up to 1.80 to 1.90 kg,
after which any increase in BiW did not result in an increase in growth rate. In the
population of pigs below this break point value, WW was the most critical risk factor
for predicting lifetime growth whereas BiW was the critical factor for the interval WW
to FW when plotted against WW. The absence of any breakpoint for IW indicates a
linear relationship for which any increase in IW will result in an increase in growth rate
to FW.
Pigs that have undergone a period of limitation in their growth, for example during feed
restriction, may compensate when normal feeding is restored to reach a similar BW to
unaffected pigs as long as the previous management has not been too severe (Kyriazakis
and Emmans, 1991; Kyriazakis et al., 1991). Similarly, it has been postulated that light
BiW piglets can exhibit varying degrees of catch up growth to meet or exceed the
slaughter weights of littermates with heavier BiW (Handel and Stickland, 1988), the
extent of which is reliant upon the number of muscle fibres present at birth (Handel and
Stickland, 1988; Dwyer et al., 1993). However this finding is not consistent in the
literature, as Rehfeldt and Kuhn (2006) and Beaulieu et al. (2010) have stated that light
BiW piglets were more likely to exhibit poor growth rates and lighter BW at successive
intervals, pre- and postweaning, therefore not meeting the BW of their heavier
37
littermates. The results of this study support the view that piglets with lower BiW may
reduce the deficit in their growth during the postnatal period; in addition almost one-
half of these pigs were able to meet or exceed the average FW of the population. When
looking specifically at light BiW piglets for each dataset, piglets with the lightest BiW
were more likely to increase a weight category by FW than those heavier pigs. Whilst it
is inevitable that those piglets in the heaviest and lightest BW categories can only
change category in 1 direction (i.e., increase or decrease, respectively), these results do
show a degree of catch up growth is occurring in light BiW piglets, and in some cases
these piglets were able to catch up to heavier littermates by FW.
3.5 Conclusions
Our results provide a better understanding of the factors affecting postnatal growth in
the pig. The implications of this study are that, although a number of risk factors are
associated with poor growth in pigs, both light BiW and WW result in poor growth to
finishing. However some of these small pigs appear to have the capacity to compensate
for low BW at birth, suggesting that there needs to be renewed focus on interventions in
the earlier stages of production to maximise postnatal growth of low BW pigs.
38
Chapter 4: A high nutrient specification diet at 9 weeks of age does not
improve the performance of low birth weight pigs
4.1 Introduction
Variability in BW of pigs is an important factor for both producer and processor in
detracting from maximum return (Patience and Beaulieu, 2006). Low birth weight pigs
contribute to this variation by exhibiting poor growth rates (Poore and Fowden, 2004;
Gondret et al., 2005; Rehfeldt et al., 2008; Douglas et al., 2013), likely as a result of
restriction in utero (Widdowson, 1971; Wu et al., 2006; Wang et al., 2008), which is
exacerbated by competition from heavier pigs for limited resources (Algers and Jensen,
1991; English, 1998; Lay et al., 2002).
As a consequence of restricted nutrition in utero, the body composition of LBiW piglets
differs from heavier littermates (Rehfeldt and Kuhn, 2006), and they may be able to
meet the BW of heavier littermates (Paredes et al, 2012; Douglas et al, 2013). It is
therefore possible that these pigs may benefit from a higher nutrient specification diet
rather than a diet that targets the ‘average’ pig (Wellock et al., 2004; Kyriazakis and
Houdijk, 2007). In fact, postnatal growth of LBiW pigs often results in a tendency to
deposit more adipose tissue (Bee, 2004; Rehfeldt and Kuhn, 2006), which may be a
reflection of such ‘inappropriate’ feeding. These principles are exploited in human
LBiW infants fed a high nutrient formula during the first year of life to increase BW
(Kashyap et al., 1994; Premji et al., 2006), with similar results observed in
supplementary milk feeding of pigs (Morise et al., 2011; Jamin et al., 2012; Han et al.,
2013).
Therefore, we hypothesized that LBiW pigs would respond to feeding of an improved
nutrient specification diet which is higher in amino acids: energy, in a manner similar to
NBiW pigs which are of similar weight for age due to experiencing a period of feed
restriction post weaning. The aim of the experiment was to determine the performance
responses of pigs with different weight for age, resulting from different prenatal or
postnatal growth, given either a high or standard amino acid: energy ratio diet.
39
4.2 Materials and method
4.2.1 Experimental design
The experiment was designed as a 3 x 2 factorial with 6 replicates. Treatments
comprised 3 BW categories (N = NBiW (1.6 to 2.0 kg), NR = NBiW but fed restrictedly
from d 49 to 63 and L = LBiW (≤ 1.25 kg)) and two diet specifications (HP = high
amino acid: energy ratio (supplying 14 g SID lysine/MJNE), SP = standard amino acid:
energy ratio (supplying 11 g SID lysine/MJNE)) from d 63 to 91 of age. The experiment
was conducted at Cockle Park Farm, Newcastle University and was approved by the
Animal Welfare and Ethics Review Board at the University.
4.2.2 Farrowing, lactation and weaner management
At the beginning of period 1 (birth to d 49), a total of 180 crossbred pigs (dam was
Large White x Landrace cross and sire was Hylean synthetic, Hermitage Seaborough
Ltd., Devon) from 6 consecutive farrowing batches were selected based on BiW. Each
farrowing batch consisted of approximately 17 litters, and from these L and N piglets
were selected. Piglets which did not meet the weight requirements were fostered onto
non experimental sows. Within the first 12 h after birth pigs were teeth clipped,
weighed and individually ear tagged for identification. Morphometric measurements
were also taken during this period: crown-rump length (CRL), snout-ears length and
abdominal and cranial circumference.
Efforts were made during the lactation period to maximize the growth of all pigs by
reducing limiting factors such as competition from heavier littermates and poor milk
supply. During the first 24 h all piglets selected for trial were cross fostered into a litter
according to their BiW i.e. 2 sows for N (one will later be restricted to form NR) and
one for L. This procedure is commonly used on farms in the UK, where piglets are
regrouped to create more uniform litters. This ensures that the teats on the sow are
accessible for the size of the piglets. An effort was made to have an equal number of
piglets of each sex and from at least 3 different birth litters in each cross fostered litter.
All sows used were first or second parity sows to ensure small piglets could access the
teats; each sow was also checked to ensure there were sufficient functional teats to
support the litter. All litters were offered supplementary Faramate milk (Volac
International Ltd, Orwell, Hertfordshire; Protein = 22%, oil = 14%, ash = 7.5%, fibre =
0%) ad libitum from birth to weaning (~28 d). This was provided in a metal dish and
40
was refilled twice a day as needed; it was prepared by hand mixing milk powder with
warm water. Piglets received commercial creep feed from day 10 onwards; the creep
feed was placed once a day on the floor of the heated creep area and was the same as the
starter 1 diet offered at weaning (23% crude protein (CP), 16.0 MJ/kg DE, 1.43% total
lysine).
Pigs were weaned on 28 d, where they were transferred to nursery accommodation with
plastic slatted floors and kept in their pre-weaning litters. Aluminium ear tags were
removed and replaced with plastic weaner tags. The temperature in the nursery
accommodation was 26 °C and was reduced by 0.2 °C/d to a minimum of 22 °C over a
period of 20 d. Each pen had a feeder with 3 spaces and a separate nipple drinker, and
all pigs had ad libitum access to feed and water. Pigs were fed a 3 stage starter diet
regime (Primary Diets, Ripon, North Yorkshire; diet 1 = 23% CP, 16.0 MJ/kg DE,
1.43% total lysine, diet 2 = 22% CP, 15.25 MJ/kg DE, 1.33% total lysine and diet 3 =
21.7% CP, 15.0 MJ/kg DE, 1.28% total lysine) with fixed amounts per pig and given
sequentially lasting 2 to 3 wk. Piglets were individually weighed twice a week from
birth to d 49 and FI per litter was measured from d 28 to 49.
4.2.3 Experimental management
Treatments did not start until d 49 onwards to ensure that all pigs had recovered from
any post-weaning growth check. At d 49 pigs were moved to experimental
accommodation, where each litter was split to form 2 treatment groups of 5 pigs each
(balanced for sex and litter of origin using SAS Proc plan to randomly allocate pigs to
treatments); any additional pigs were removed from the experiment. The
accommodation consisted of partly slatted concrete floors; each pen provided 0.96
m2/pig and had a feeder with 5 spaces and a separate nipple drinker which was located
over the concrete slats. A thermostatically controlled heating and fan ventilation system
maintained the room temperature between 19 and 21 °C, which was monitored daily
using a max-min thermometer.
In period 2, from d 49 to 63, the NR pigs received restricted amounts of feed (600 g/d
per pig) with the remaining N and L groups fed ad libitum the same commercial weaner
diet (A-One Feed Supplements, Thirsk, North Yorkshire; 20.55% CP, 14.46 MJ/kg DE,
1.28% total lysine). The aim was for NR and L pigs to have the same BW by d 63. The
41
amount of feed given to NR pigs was calculated using previous performance data
(Average daily feed intake (ADFI) and FCE) of similar weight pigs from Cockle Park.
For period 3, from d 63 to 91, groups within litter were randomly allocated a high (14 g
SID lysine/MJNE) or a normal amino acid: energy (11 g SID lysine/MJNE) grower diet
for their age, offered ad libitum for 4 wk. Table 4.1 reports the composition of the
experimental diets used.
From d 49 to 91, pigs were individually weighed twice a wk, on a Monday and
Thursday morning. Feed intake was determined by manually recording the total feed
given for each 3 or 4 d period, and the feed refusals prior to the next weighing. With
these measurements, ADG was calculated for individual animals, and ADFI and FCE
were calculated for pens. Both ADG and ADFI were scaled to BW to allow for
comparisons and account for the fact that pigs were of different size (Kyriazakis et al.,
1991). Various exponents methods were tested to scale ADG and ADFI to BW (e.g.
BW0.75
,BW0.66
), but, as these had no effect on the conclusions drawn; here we report the
outcomes per unit BW. The scaled ADG (SDG) was calculated as ADG/kg BW whilst
the scaled ADFI (SFI) was ADFI/kg BW. The BW used to scale the ADG/ADFI was
the weight at the start of the specific period in question, for example when calculating
the SDG for d 63 to 91, the ADG for this period was divided by the BW on d 63.
Morphometric measurements taken at birth were used to calculate the relative CRL
(CRL/kg) and PI (BW/CR3).
42
Table 4.1 Diet composition and chemical analysis for the dietary treatments, SP and
HP, offered from d 63 to 91 of age. The two feeds contained different amino acid:
energy ratios
Diet
Item SP HP
Ingredient, g/kg
Wheat 57.5 46.1
Micronized barley flakes 20.0 20.0
Soya bean meal 13.4 21.2
Pig finisher premix2 0.25 0.25
Full fat soya (Soycomil) 4.75 7.50
L-lysine-HCL 0.49 0.55
DL-Methionine 0.13 0.23
L-Threonine 0.18 0.22
L-Tryptophan 0.02 0.03
Vitamin E 0.01 0.01
Limestone flour 0.28 0.33
Dicalcium phosphate 0.95 0.85
Salt 0.22 0.33
Sodium bicarbonate 0.17 0.00
Binder (Lignobond DD)
Soya Oil
0.63
1.04
0.63
1.82
Analysed composition2 % as fed
NE, MJ/kg3
Ash
10.1
3.70
10.1
4.10
Crude fibre 2.50 2.80
Crude protein 16.8 21.3
Total lysine 1.09 1.58
Methionine 0.37 0.56
Oil 3.92 4.65
Moisture 11.9 11.3 1 Provided per kg of complete diet: 9,000 IU of vitamin A, 2,000 IU of vitamin D3, 35
IU of vitamin E, 2 mg of vitamin K, 1.5 mg of vitamin B1, 4 mg of vitamin B2, 2.5 mg
of B6, 15 µg of vitamin B12, 11 mg of pantothenic acid, 15 mg of nicotinic acid, 50 µg
of biotin, 0.5 mg of folic acid, 15 mg of CU (CUSO4), 1.5 mg of Iodine (KI, Ca(IO3)2),
100 mg of Fe (FeSO4), 35 mg of Mn (MnO), 0.25 mg of Se (BMP-Se), 100 mg Zn
(ZnSO4). 2 Analysis performed by Sciantec Analytical Services Ltd (North Yorkshire)
3 Values estimated using raw material matrix (Primary Diets., Melmerby, UK)
43
4.2.4 Statistical analysis
All performance data was tested for normality using the Univariate procedure of SAS
version 9.2 (SAS Institute Inc., Cary, NC) and was normally distributed. Sex (entered as
a proportion of females in each pen or litter depending upon the period being examined)
was included as a factor in all preliminary models but was not significant so omitted for
subsequent analysis. Data was blocked by sow to account for litter effects. Treatment x
sow interactions were added to all preliminary models, but were not significant.
Differences were considered significant at < 0.05 and reported as a tendency towards
statistical significance at < 0.10. Data are presented as least square means.
For d 1 to 49 only, the effect of BiW on performance indicators were analysed using
repeated measures ANOVA using Proc Mixed of SAS. Suckling group (‘N’ or ‘L’) and
time (d 1 to 49) were added as factors. The experimental unit from d 1 to 49 was the
litter or suckling groups L or N (11 to 12 piglets). For d 49 to 91, the effect of BW
category (N, NR and L), diet specification in period 3 (high or standard amino acid:
energy diet) and time were analysed using a repeated measures ANOVA using Proc
Mixed. The experimental unit was the pen (5 pigs). Body weight category, diet
specification and time were added as factors.
4.3 Results
4.3.1 Performance in period 1, d 1 to 49
The effects of BW categories NR, N and L on weight, ADG, ADFI and FCE from birth
to d 91 are shown in Table 4.2. At birth, L pigs had an average weight of 1.02 kg (SD
0.152) whilst N pigs averaged 1.87 (SD 0.103). Focusing specifically on the
performance of pigs during lactation (d 1 to 28), there was no difference in the ADG of
all groups (P > 0.05). When pigs were weaned at d 28, L pigs still had a lower BW than
pigs in the other weight categories (P < 0.001), with over 1.5 kg difference in average
WW when L pigs were compared to N. In the initial period following weaning (d 28 to
49), a lower ADG of L pigs in comparison with N was seen (P < 0.001); by d 49 there
was a 3 kg difference in the BW of L and N pigs. Average daily feed intake measured
for the litters from d 28 to 49 showed that L pigs ate significantly less compared to N
pigs (Table 4.2).
44
In contrast, SDG of L pigs exceeded that of N until weaning, but from d 28 to 49 there
was no significant difference between these two groups (Fig.4.1). Scaled feed intake
also showed that L pigs exceeded the intake of N pigs from d 28 to 49, relative to their
body size (P < 0.001) (Fig.4.2).
At birth, L pigs had a shorter CRL of 23.5 cm (SD 2.11) compared to 28.7 cm in N (SD
2.51) (P < 0.001). However L pigs had a higher CRL/BW (23.4 cm/kg, SD 2.91) than N
pigs (15.4 cm/kg, SD 1.04) (P < 0.001). Ponderal index data showed that L pigs had a
significantly lower PI (86.5 kg/m3, SD 13.2) than N pigs (102.1 kg/m
3, SD 11.1) (P <
0.001).
4.3.2 Performance in period 2, d 49 to 63
On d 49, BW for L pigs differed significantly from N and NR, which did not differ from
each other (Table 4.2). Pigs fed restrictedly during this period (NR) had a lower ADFI
and ADG as expected; this resulted in them weighing the same as L pigs by d 63, whilst
N pigs were significantly heavier. L pigs grew at a significantly slower rate than N pigs,
however they ate the same as N Pigs (absolute FI) (P < 0.001). Figure 4.1 and 4.2
demonstrate that for both SDG and SFI, L pigs exceeded N pigs (P < 0.001) during this
period. There was no significant difference in the BW, ADG or FCE of the different
treatment groups prior to starting the nutritional treatments (data not shown).
45
Table 4.2 The effect of body weight categories on the performance of pigs from birth to d 91. NR were normal birth weight pigs but fed
restrictedly (between d 49 to 63), N were normal birth weight pigs and L was low birth weight pigs.
Body Weight category1
Item NR N L SEM P-value2
BW, kg
D1 1.86A 1.88
A 1.02
B 0.259 < 0.001
D 28 8.38A 8.36
A 6.84
B 0.345 < 0.001
D 49 17.6A 17.6
A 14.4
B 0.525 < 0.001
D 63 21.2A 25.8
B 21.8
A 0.610 < 0.001
D 91 44.2A 46.6
A 40.8
B 1.048 < 0.001
ADG, kg/d
D 1 to 28 0.239 0.238 0.213 0.0123 0.452
D 28 to 49 0.437A 0.440
A 0.362
B 0.0138 < 0.001
D 49 to 63 0.261A 0.595
B 0.529
C 0.0183 < 0.001
D 63 to 91 0.819A 0.742
B 0.684
B 0.0278 < 0.001
ADFI, g/kg
D 28 to 49 483A 472
A 430
B 17.7 0.023
D 49 to 63 600A 925
B 927
B 36.6 < 0.001
D 63 to 91 1279A 1224
A 1259
A 36.2 0.378
FCE
D 28 to 49 0.971 1.033 0.905 0.0588 0.597
D 49 to 63 0.423A 0.666
B 0.552
C 0.0303 < 0.001
D 63 to 91 0.660A 0.626
A 0.554
B 0.0195 < 0.001
1 A, B, C Within a period, means with different superscripts differ (P < 0.05)
46
Figure 4.1 The effect of body weight (BW) category (N, NR and L), on the scaled ADG
(SDG) of pigs from d 28 to 91. Weight category: N were normal birth weight pigs (1.6
to 2.0 kg), NR was normal birth weight but fed restrictedly (d 49 to 63) and L was low
birth weight (≤ 1.25 kg). SDG was calculated as the ADG for specific period/BW at the
start of that period. Error bars represent the pooled SEM.
0.000
0.010
0.020
0.030
0.040
0.050
0.060
0.070
SDG, kg.d-1kg
BW-1
Age, days
N
NR
L
47
Figure 4.2 The effect of body weight category (N, NR and L), on the scaled feed intake
(SFI) of pigs from d 28 to 91. Weight category: N were normal birth weight pigs (1.6 to
2.0 kg), NR were normal birth weight but fed restrictedly (at day 49 to 63) and L were
low birth weight (≤ 1.25 kg). Scaled feed intake (SFI) was calculated as the average
daily feed intake of a specific period/BW at the start of period. Error bars represent the
pooled SEM
4.3.3 Performance in period 3, d 63 to 91
The effects of BW category and diet specification on the live weight, ADG, ADFI and
FCE are shown in Table 4.3. For d 63 to 91 there was a significant difference in ADG
between BW categories; however there was no effect of diet specification on the
performance of pigs or any BW category x diet specification interaction.
Post d 63, there was a significant difference in the ADG of the BW categories; at all
stages examined L pigs grew at a slower rate than both N and NR pigs from d 63 to 91
(P < 0.001). Conversely, NR pigs performed significantly better than both N and L pigs
once the feed restriction was stopped on d 63. NR pigs had a higher ADG compared to
0.0000
0.0100
0.0200
0.0300
0.0400
0.0500
0.0600
0.0700
SFI, kg.d-
1kgBW-1
Age, days
N
NR
L
48
other treatments (P < 0.001) and by d 91 did not significantly differ in weight from N
pigs (P < 0.001). There was no effect of BW category, diet specification or an
interaction between the two, on the absolute ADFI of pigs for d 63 to 91. From d 63 to
91 L pigs had a reduced FCE in comparison to N and NR pigs. In no case did the diet or
interaction between BW category and diet have a significant effect on the FCE.
When taking into account the ADG relative to the BW of the pigs (SDG), L and N pigs
grew at a similar rate, whilst NR pigs had greater SDG in comparison (P < 0.001). In
contrast, NR and L pigs had greater SFI than N pigs from d 63 to 91 (P < 0.001).
49
Table 4.3 The effect of body weight category and diet specification (standard (SP) or high (HP) amino acid: energy ratio) on the performance of pigs
from d 63 to 91. NR was normal birth weight but fed restrictedly (d 49 to 63) pigs, N was normal birth weight pigs and L was low birth weight pigs
Treatment (Body weight category/ diet specification) Significance
NR SP NR HP N SP N HP L SP L HP SEM Weight category Diet
specification
Weight category x diet
specification
ADG, kg/d 0.785 0.854 0.730 0.753 0.683 0.684 0.0276 < 0.001 0.485 0.376
ADFI, g/kg 1,300 1,257 1,214 1,234 1,261 1256 37.4 0.378 0.246 0.357
FCE 0.624 0.697 0.618 0.633 0.551 0.557 0.19 < 0.001 0.481 0.573
50
4.4 Discussion
We investigated the ability of pigs of different weight for age to exhibit catch up growth
when given access to diets which differed in amino acid: energy content. It was
hypothesized that LBiW pigs given access to a diet of a higher nutrient specification
(diet HP), would be able to show greater ADG in comparison to those fed a standard
diet targeting the ‘average pig’ (diet SP). This was based on the principles of Kyriazakis
and Emmans (1991, 1992) who suggested that pigs previously delayed in their growth
are able to exhibit a higher degree of catch up growth when they are fed a diet greater in
protein: energy ratio. LBiW pigs may have altered body composition in relation to
heavier littermates both at birth and slaughter, with a higher percentage of adipose tissue
(Bee, 2004; Rehfeldt and Kuhn, 2006; Rehfeldt et al., 2008). Therefore it was expected
that they would benefit from a diet which is higher in amino acid: energy ratio, allowing
them to exhibit catch up growth, as has been previously demonstrated to be possible
(Paredes et al., 2012; Douglas et al., 2013; Pardo et al., 2013b). However, pigs with
LBiW given access to such an ‘improved’ diet in period 3 showed no improvement in
performance.
By the end of the experiment NR pigs had partially caught up with N pigs. This was
consistent with previous work which has demonstrated that pigs have the potential to
compensate for ‘moderate’ postnatal stunting if subsequently fed an adequate diet for a
sufficient period of time (Lynch et al., 1998). While the previously restricted pigs did
not increase their absolute FI relative to N controls, an increase in SFI was observed. It
seems that the extent of catch up growth of the NR pigs did not depend on the
composition of the diet offered post restriction. Kyriazakis and Emmans (1992) and
Stamataris et al. (1991) have both suggested that this will depend on the consequences
of restriction on the body composition of pigs. Pigs that have been delayed in their
growth, but also have a higher lipid to protein ratio in their body are expected to benefit
more from a diet of a higher amino acid specification. These results indicate that NBiW
pigs that experience a period of involuntary feed restriction (e.g. reduction in the
amount of feed provided), are likely to catch up when the non-limiting conditions are
resumed, even when not fed a high specification diet.
51
In contrast, the results suggest that, unlike NBiW pigs, LBiW pigs cannot catch up with
heavier littermates, even on a higher specification diet. It was hypothesized that, when
given an appropriate nutritional environment, irrespective of the reasons that led to the
reduced BW (i.e. LBiW or feed restriction post weaning), pigs would be able to meet
the BW of heavier pigs. The absence of any benefit of the LBiW when fed the high
specification diet was surprising and could be the result of several reasons.
Firstly, it must be considered whether these pigs are physiologically capable of
improving their growth. There is a wealth of literature which focuses on the uterine
environment and changes in the physiology of LBiW pigs, as a result of reduced
nutrition in utero. In agreement with previous literature (Nissen and Oksbjerg, 2011) we
found that LBiW pigs had a significantly lower CRL and a higher relative CRL
(CRL/BW) than NBiW pigs. When also considering the lower PI associated with LBiW
pigs this indicates disproportionate body size at birth which is likely a result of
restriction in utero (Poore et al., 2002). It is commonly thought that these pigs are born
with a reduced capacity for growth (Foxcroft et al., 2006) and it has been documented
that there are a reduced number of muscle fibres in LBiW pigs compared to heavier
littermates (Powell and Aberle, 1980; Handel and Stickland, 1988; Dwyer et al., 1994;
Rehfeldt and Kuhn, 2006). As the number of these fibres is fixed in utero, muscles will
only grow by hypertrophy, which consequently may limit growth. These suggestions,
however, are not supported by the recent analysis of large data sets that show that such
pigs are indeed capable of compensatory growth (Paredes et al., 2012; Douglas et al.,
2013).
Secondly, it has been suggested that the FI of LBiW pigs is a limiting factor in their
growth (Gondret et al., 2005), although the results presented here dispute this argument.
With the exception of the immediate period following weaning (d 28 to 49), there was
no difference in the absolute ADFI of LBiW and NBiW pigs. This supports the
suggestion of Gondret et al (2006) who concluded that LBiW and high BiW pigs had
similar feed consumption during the grower and finisher stages. In addition scaled feed
intake (SFI) comparisons showed that, relative to their body size, LBiW pigs were
matching or even exceeding the intake of NBiW pigs throughout all periods examined.
This was in agreement with the consistently higher relative FI by LBiW reported by
Krueger et al (2013).
52
Thirdly, the experimental method used may have prevented the light pigs from
exhibiting catch up. The ability of the animal to overcome its growth constraints will
vary between individuals and is dependent on the composition of the food it receives
and the environment it is kept in, as well as the current state of the pig following
nutritional limitation (Kyriazakis and Houdijk, 2007). In this experiment animals were
given access to the diet with a higher ratio of amino acids: energy from 9 wk of age.
Between weaning and this age they received a diet that was based on age rather than
weight, so possible limiting their growth potential. It is possible that intervention earlier
on may show different results.
Whilst LBiW pigs did not show an improvement in ADG when given a diet higher in
amino acid: energy, it is important to dissect the overall performance of these pigs
during the experiment. During the pre-weaning period, LBiW pigs had similar ADG to
heavier piglets, contradicting a number of studies which observed light pigs exhibiting
poorer growth rates in this period (Dwyer et al., 1994; Quiniou et al., 2002). Inevitably
differences in experimental method will affect the outcomes observed, given the strong
influence of pre-weaning competition on piglet growth. For example, in this study all
piglets were given access to supplementary milk as well as being grouped in litters with
similar sized littermates. Competition for access to teats during suckling and
consequently low milk consumption is considered a limiting factor for pre weaning
growth (Campbell and Dunkin, 1982). Reducing this disadvantage of LBiW pigs by the
addition of milk is likely to enhance their growth and reduce mortality (Azain et al.,
1996; Wolter et al., 2002). Whilst the absolute growth rate of N and L pigs from d 63 to
91 were not statistically different, L pigs still grew considerably slower. However when
we consider these data on a per kg of BW basis (SDG), LBiW pigs actually grew at the
same rate or better than the heavier pigs N. This suggests that LBiW pigs can exhibit
growth rates not extremely dissimilar from NBiW pigs given the right conditions.
It has been suggested that LBiW pigs are less efficient than their heavier counterparts,
with poorer gain to feed ratios in LBiW pigs having been observed (Roeder and Chow,
1972; Gondret et al., 2005). With the exception of the period following weaning (d 28
to 49), the results presented here support these findings. When the data for both ADG
and ADFI of the pigs are considered, despite the LBiW pigs eating at least the same or
more than the NBIW pigs their gains are slightly less (although not significantly).
53
Whilst it is apparent that the appetite of LBiW pigs is not affected (post 7 wk) by
restriction in utero it is possible other processes may be influenced.
4.5 Conclusions
The results suggest that a diet higher in amino acids: energy ratio at 9 weeks of age does
not improve the performance of LBiW pigs. These pigs consume a similar amount of
feed, but as they sometimes exhibit reduced BW gains, they are thus less efficient than
their heavier counterparts. However, they are capable of exhibiting similar gains to
NBiW pigs at certain periods in the production cycle, and this needs to be investigated
as a possible method for improving their growth and reducing weight variability at
slaughter.
54
Chapter 5: Management strategies to improve the performance of low
birth weight pigs to weaning and their long term consequences
5.1 Introduction
Increases in litter size in recent years have resulted in significantly more piglets born
with LBiW (Beaulieu et al., 2010; Baxter et al., 2013; Rutherford et al., 2013).
Subsequent growth of these piglets is often below average, and at slaughter these pigs
can weigh significantly less than their pen mates. In order to maximize the growth of
these LBiW pigs, and reduce variability, there needs to be renewed focus on the earlier
stages of production (Pluske et al., 2005; Douglas et al., 2013).
Weight gain of pigs during the preweaning stage varies significantly, given the many
influential factors. Piglets are reliant on milk from the sow and, during lactation, sibling
competition may have a major effect on survival and growth (Algers and Jensen, 1991;
English, 1998; Lay et al., 2002). This is likely to affect small pigs the most, often
exacerbating the difference in BW by weaning, leaving these piglets further
disadvantaged.
Presently there are few treatments which can improve the growth of LBiW pigs during
lactation or at any other stage during production. Providing piglets with supplemental
milk replacer during lactation can improve WW (Kim et al., 2001; English and Bilkei,
2004; Morise et al., 2011), although the benefits for growth to slaughter are
inconclusive. It has also been suggested that LBiW pigs are at a competitive
disadvantage when raised with heavier littermates, therefore they may perform better in
litters with less weight variability (English, 1998).
The objective of this study was to determine the effect of littermate weight and milk
supplementation during lactation on the growth performance of LBiW pigs to slaughter
weight. It was hypothesized that there would be an interaction between littermate
weight and milk supplementation, with LBiW pigs in mixed litters being more likely to
benefit from milk supplementation due to greater competition from heavier littermates
for limited resources. The long term effects of these treatments were also evaluated.
55
5.2 Materials and Method
5.2.1 Experimental design
The experiment was conducted at Cockle Park Farm, Newcastle University and was
approved by the Animal Welfare and Ethics Review Board at the University. The
experiment was designed as a 2 x 2 factorial with 6 replicates. A total of 265 crossbred
piglets (dam was Large White x Landrace cross and sire was Hylean synthetic,
Hermitage Seaborough Ltd., Devon, UK) were cross fostered onto 24 sows. Treatments
were litter composition (L = low birth weight pigs only or MX= Low birth weight and
normal birth weight pigs) and provision of milk supplement from d 1 to 28 (Y= Yes, N
= No). There were 6 replicates of each treatment. The experimental unit was the litter
mean of all LBiW pigs.
5.2.2 Animal management
Sows were farrowed on a 3 wk cycle in individual farrowing crates which were
equipped with a feeder and drinker for the sow. A total of 6 batches were used, each
batch was a full replicate. They were allowed to farrow normally at term over a four day
period (Monday to Thursday); sows that had not farrowed within this period were then
induced on Thursday by injection of a prostaglandin analogue. All piglets were teeth
clipped within the first 12 h of birth, and were tail docked and given an iron injection at
d 3. The temperature in the farrowing house was maintained at 21°C by a centrally
controlled heating and ventilation system and an infra red heat lamp was located in the
creep area for the piglets to provide a microclimate during the lactation period. The
nutrition of the sows was standardised across all treatments, with a home milled meal
fed prior to and during lactation (18.5% CP, 13.98 MJ DE, 0.95% total lysine). Sows
were fed 2.0 kg/d prior to farrowing; this was then increased by increments of 0.5 kg/d
until they were fed 10 kg/d. From d 10 onwards, a small amount of pelleted creep feed
for the piglets was placed on the floor of the heated creep area once a day and was the
same as the starter 1 diet fed at weaning (23% CP, 16.0 MJ DE, 1.43% total lysine).
Pigs were weaned at 28 d, when they were transferred to controlled environment
nursery accommodation with plastic slatted floors and housed in their pre-weaning
litters. Pigs were vaccinated for Mycoplasma hyopneumoniae and Porcine Circovirus
Type 2 (Inglevac Mycoflex, Boehringer Ingelheim, Germany). The initial temperature
in the nursery accommodation was 26 °C and was reduced by 0.2 °C/day to a minimum
56
of 22 °C. Each pen had a multi-place feeder with 3 spaces and a nipple drinker; all pigs
had ad libitum access to feed and water. Pigs were fed a standard 3 stage commercial
diet regime from d 28 for approximately 2 to 3 wk (Starter diet 1 = 23.0% CP, 16.0 MJ
DE, 1.43% total lysine; diet 2 = 22.0% CP, 15.25 MJ DE, 1.33% total lysine; and diet 3
= 21.7% CP, 15.0 MJ DE, 1.28% total lysine). After this period all pigs were fed the
same, home milled meal ad libitum (20.5% CP, 14.82 MJ DE, 1.28 % total lysine). At
approximately 10 wk of age pigs were transferred to a separate controlled environment,
fully slatted grower accommodation on site, where they were fed the same home mixed
‘grower’ diet (20.04% CP, 13.98 MJ DE, 1.20% total lysine). At approximately 16 wk
of age they were moved again to a fully slatted finishing building where they were fed a
purchased ‘finisher’ diet ad libitum (19.0% CP, 13.64 MJ DE, 1.10 % total lysine) until
slaughter at approximately 140 d. After pigs were moved to the grower accommodation
they were randomly mixed by litters according to farm protocol. In both grower and
finisher accommodation, each pen had a multi-space feeder with 3 spaces and a nipple
drinker; all pigs had ad libitum access to feed and water.
5.2.3 Experimental procedures
Within 12 hours of birth all piglets were weighed and those selected for the experiment
were ear tagged for identification. Pigs were individually identifiable at all stages of the
experiment. Low birth weight piglets (LBiW) were classified as weighing ≤ 1.25 kg and
NBiW piglets as weighing 1.6 to 2.0 kg (Douglas et al., 2013) at birth; piglets that did
not meet these weight criteria were cross fostered onto non experimental sows.
Morphometric measurements were also taken at birth: CRL, snout-ears length and
abdominal and cranial circumference, and used to calculate the relative CRL (CRL/kg)
and PI (BW/CRL3).
To create the experimental litters, all piglets were randomly assigned to a litter within
24 h after birth. Where possible, each experimental litter contained an equal number of
piglets of each sex. To ensure there was no litter of origin effect, cross fostered litters
consisted of pigs from at least 4 different birth litters, with no more than 3 piglets from
the same litter. Litter size was set at 11 or 12 piglets, depending on the number of
suitable piglets available in a batch. In L litters 11 to 12 LBiW pigs were grouped
together, whereas MX litters consisted of 5 to 6 LBiW and 5 to 6 NBiW piglets. All
sows used were first or second parity to ensure small piglets could access their teats.
57
Each sow was also checked to ensure there were sufficient functional teats to support
the litter size allocated.
Once the experimental litters were set, they were randomly assigned within batch to the
milk supplementation treatment. Half of the litters were given access to supplementary
milk (S), and half were not (N). A feeder containing supplementary milk for the piglets
was added to the pen of S litters from 24 h after birth. This comprised a dish attached to
the slats at the rear of the pen, which was filled twice a day or as needed. Milk
consumption was monitored throughout the day to ensure there was always milk
available. If milk was found to be contaminated then it was discarded and fresh milk
was added. Any discarded milk was measured to ensure accurate estimation of milk
intake. To minimise milk spillage a small metal bowl 25 cm diameter and 3.5 cm depth
was used for the first 10 d; this was then replaced with a larger plastic bowl of 37.5 cm
diameter and 6.5 cm depth. Milk was prepared by hand mixing 150 g commercial milk
powder (Faramate, Volac, Royston, UK; Protein = 22%, oil = 14%, ash = 7.5%, fibre =
0%) with 1 L of warm water. Piglet snouts were dipped in the milk for training on two
consecutive days.
Daily milk intake was recorded for each litter by measuring the milk added and refused;
in addition cameras were set up to record piglet behaviour at the milk dish. Piglets were
observed for signs of diarrhoea (nutritional scours) related to the milk supplement and
other illness. Any piglets that exhibited diarrhoea were treated with 0.5 to 1.0 ml of
Norodine (Norodine 24 solution for injection, Norbrook, Corby, UK) depending on the
BW; if 3 or more piglets in a litter showed symptoms, the whole litter was treated. If
piglets did not reach 4 kg BW by d 28 then they were not weaned and were removed
from trial.
From d 1 to 28, pigs were individually weighed twice a week, on a Monday and
Thursday morning. From d 28 to 49, pigs were weighed once a week on a Thursday
morning. Additional weights were taken at 100 d and a final weight was taken on the
day before slaughter (approximately d 143). With these measurements, ADG was
calculated for individual animals and treatment groups, and average daily milk intake
(ADMI) for litters. The CV of BW was calculated for individual litters.
58
5.2.4 Behavioural observations
Digital video cameras were used to record piglet behaviour. All litters with
supplementary milk (both L and MX) were observed on d 13, 20 and 27, in order to
cover the period at which lactation yield plateaus (Nielsen et al., 2002; Hansen et al.,
2012) and piglets should increasingly seek extra nutrition from the supplementary milk.
On observation days, piglets were weighed and marked on the back with individual
markings (different symbols and colours). At approximately 9 am the cameras were
then turned on and fresh milk supplement was added to the milk bowl. The videos were
left on for 24 hours and turned off the following morning at approximately 9 am. From
4pm the lights in the farrowing house were switched off (with the exception of the heat
lamps in the creep area) so an additional light was added above the milk dish to allow
the cameras to record the piglets. Subsequently a continuous record of each pig’s
behaviour for an 8 hr period from 8am until 4 pm was obtained using the behavioural
research software, Observer 5.0 (Noldus Information Technology, Wageningen, the
Netherlands) to quantify the frequency and duration of the following behaviours:
1. Drinking supplementary milk – defined as a piglet being at the bowl with its
head down for longer than 3 s, and
2. Suckling – the start of the suckling bout was defined as when 8 or more piglets
gather at the sow udder and begin massaging. The suckling bout was finished
when 8 or more piglets had stopped massaging and moved away from the udder,
or the sow moved position and therefore terminated the suckling bout
5.2.5 Statistical analysis
Growth performance (ADG) was summarised for the following periods: d 1 to 14, d 14
to 28, d 28 to 49, d 49 to 100 and d 100 to 143. All performance data was tested for
normality using the Univariate procedure of SAS version 9.2 (SAS Institute Inc., Cary,
NC) and was normally distributed. For behavioural data, normality testing showed that
the data were skewed, so they were transformed (log or square root) and then results
back transformed for presentation. Statistical analyses were conducted at a significance
level of 5% and data presented as least square means. Data for all analyses were blocked
by farrowing batch (6 batches). Treatment x batch interactions were added to all
preliminary models, but were not significant and therefore omitted from subsequent
analysis.
59
A chi-square test was used to compare the effects of litter composition (L or MX) and
milk supplementation (S or N) on the reason for removal of pigs from trial and also for
the occurrence of diarrhoea. The effect of litter composition on milk intake was
estimated using a one way ANOVA using Proc mixed of SAS. For both analyses, the
litter was the experimental unit.
For LBiW pigs in L and MX litters only, performance parameters were entered as the
dependent variables to determine the effect of treatments on the different performance
indicators. The experimental unit was the litter mean of all LBiW pigs; for L this was 11
to 12 piglets, and for MX litters this was 5 to 6 LBiW piglets. After litters were mixed
at d 70 then pens were considered the experimental unit. The effect of littermate weight
and milk supplementation on performance indicators was analysed with a repeated
measures ANOVA using Proc Mixed of SAS. Littermate weight, milk supplementation
and time were added as factors to the model. Sex was also included as a factor in the
preliminary model, but as it was not significant it was omitted from subsequent analysis.
As the number of LBiW piglets in L and MX litters varied, a weighted statistical
analysis was used. Using the weight statement of Proc Mixed, a count of the number of
LBiW piglets in each litter was added to the model as an additional variable.
Behavioural data was also analysed with a repeated measure ANOVA. The same
analysis was repeated with the following variables; number of milk feeds per 24 h, time
spent feeding (s per 24 h) and number of suckling bouts per 24 h. Time (d 13, 20 or 27)
was also included as a factor in the model. An additional model comparing the
performance of LBiW piglets in L litters and NBiW piglets in MX litters was run. The
experimental unit was the litter mean of LBiW or NBiW pigs. The model was the same
as above.
For LBiW and NBiW pigs in MX litters, performance parameters were entered as
dependent variables to determine the effect of treatments on the different performance
indicators. The experimental unit was the litter mean of LBiW or NBiW pigs. After
litters were mixed at d 70 then pens were considered the experimental unit. The effect of
littermate weight and milk supplementation on performance indicators was analysed
with a repeated measure ANOVA using Proc Mixed of SAS. Birth weight, milk
supplementation and time were added as factors to the model. Sex was included as a
factor in the preliminary model but as it was not significant it was omitted for
60
subsequent analysis. As LBiW and NBiW piglets were from the same litter and
therefore their performance was confounded, data was blocked by litter. Litter x
treatment interactions were also tested, however they were not significant so removed
from further analysis. For the behavioural data, the repeated measures ANOVA was
repeated with the following variables; number of milk feeds per 24 h, time spent feeding
(s per 24h) and number of suckling bouts per 24h. Time (d 13, 20 or 27) was also
included as a factor in the model.
5.3 Results
Litter composition and milk supplementation had no significant impact on the mortality
rate of LBiW piglets or the number of removals from the trial (Table 5.1). Treatment
with antibiotic was higher in litters with supplementary milk that those without but no
difference between MX and LBiW litters. At birth, LBiW pigs (in MX or L litters) had a
shorter CRL of 23.2 cm (SD 1.88) compared to 27.4 cm (SD 2.13) for NBiW piglets (P
< 0.001). However LBiW pigs had a significantly higher CRL/BW, (21.0 cm/kg, SD
2.81) than NBiW pigs (15.2 cm/kg, SD 1.01) (P < 0.001). LBiW pigs also had a
significantly lower PI (89.5 kg/m3, SD 10.3) than NBiW pigs (99.5 kg/m
3, SD 11.5) (P
< 0.001).
5.3.1 Performance and behaviour of low birth weight piglets (LBiW) in L or MX
litters
The effects of litter composition (L or MX) and milk supplementation (S or N) on the
BW and ADG of LBiW pigs from birth to slaughter are shown in Table 5.2. There was
no effect of milk supplementation on the BW or the ADG of LBiW pigs in L and MX
litters for all periods examined, nor was there any interaction between litter type and
milk supplementation on the performance of the piglets (P > 0.05). There was no effect
of litter type or milk supplementation on the within-litter CV of BW of pigs from birth
to d 143 (data not shown).
When considering the effect of litter type on the BW of LBiW pigs, there was no
significant difference during the earlier part of lactation (d 1 and 14). However at d 28,
LBiW pigs in L litters weighed 500g more than LBiW pigs in MX litters (P < 0.05). By
d 49 there was a 750 g difference between these pigs from L and MX litters, which had
increased to 2 kg by d 143; however neither were considered significant (P > 0.05).
61
Similarly, there was no effect of litter type on the ADG of piglets from d 1 to 14 of
lactation; however from d 14 to 28 LBiW pigs in L litters grew better than those in MX
litters (0.252 kg/d versus 0.271 kg/day; P < 0.05). Post d 28 there was no effect of litter
type on the ADG of LBiW pigs (P > 0.05).
62
Table 5.1 The reasons and numbers removed from the experiment and the number of antibiotic treatment for scours of low birth weight
(LBiW) and normal birth weight (NBiW) pigs on the experiment. Low birth weight pigs (LBiW) were either in litters with other LBiW
pigs (L) or in litters with both LBiW and ‘normal’ pigs (MX). Half of the L and the MX litters were given supplementary milk (S) from d 1
to 28 and the other half were not (N)
Treatment
L MX Significance1
Milk S N S N
No of pigs on trial
Day 1 66 67 66 66 0.951
Day 28 60 61 61 61 0.949
Day 143 59 61 61 61 0.897
Antibiotic treatment2
Reasons for removal
15 9 14 11 0.644
Scour 2 1 2 - 0.361
Lameness 2 1 2 - 0.361
Hernia - 1 - 1 -
Meningitis - - - 1 -
Found dead 2 1 1 1 0.709
Under 4 kg at day 28 1 2 - 2 0.361 1
Absence of statistics indicates there were insufficient observations for a chi-square test 2
Antibiotic treatments for scours were for individual pigs and for multiple episodes. The values only includes pigs which were treated after
being diagnosed with diarrhoea, rather than pigs which were treated as a result of 3 or more piglets in the litter having diarrhoea
63
Table 5.2 The effect of littermate weight and milk supplementation on the performance of low birth weight pigs (LBiW) pigs from d 1 to
143. The LBiW pigs were either in litters with other LBiW pigs (L) or in litters with both LBiW and ‘normal’ pigs (MX). Half of L and
MX litters were given supplementary milk (S) from d 1 to 28 and the other half were not (N)1
Littermate
weight L MX
Significance
Milk S N S N SEM Littermate weight Milk Littermate weight x Milk
BW, kg
Day 1 1.11 1.14 1.13 1.15 0.256 0.485 0.284 0.691
Day 14 3.84 3.78 3.64 3.89 0.110 0.651 0.456 0.301
Day 28 7.54 7.13 6.73 6.87 0.262 0.045 0.596 0.256
Day 49 15.5 14.5 14.2 14.3 0.523 0.101 0.460 0.345
Day 100 46.2 45.5 45.4 45.5 0.861 0.650 0.760 0.680
Day 143 74.5 74.8 72.3 73.5 1.49 0.291 0.638 0.770
ADG, kg/d
Day 1-14 0.210 0.203 0.193 0.210 0.010 0.643 0.621 0.245
Day 14-28 0.264 0.239 0.221 0.213 0.015 0.021 0.213 0.556
Day 28- 49 0.381 0.357 0.348 0.358 0.012 0.511 0.786 0.486
Day 49 to 100 0.602 0.608 0.617 0.611 0.013 0.534 0.951 0.654
Day 100 to 143 0.657 0.673 0.624 0.654 0.034 0.411 0.479 0.843 1 Piglets were grouped in litters from d 1 until d 70; following this 2 litters selected at random were combined.
64
There was no difference in the number of supplementary milk feeding episodes of
LBiW piglets between litter types for all time periods examined (P > 0.05). In contrast,
milk feeding duration on d 27 was increased in LBiW piglets in MX litters compared to
L litters (33 s versus 15.8 s; P < 0.001). There was also a difference in the number of
nursing episodes between litter types, with L litters suckling more often in an 8 h period
than MX litters for d 27 (11.6, SD 4.10 versus 8.17, SD 3.25; P < 0.001).
There was a significant difference in the milk intake of L and MX litters. Piglets in L
litters had a higher ADMI over the 28 d lactation period in comparison to those in MX
litters, with an average of 171 ml/d per pig compared to 138 ml/d respectively (P <
0.05).
5.3.2 Performance of low (LBiW) and normal birth weight (NBiW) pigs in mixed
litters
Birth weight had a significant effect on BW from d 1 to 143, with LBiW pigs weighing
significantly less at all periods examined (P < 0.001) (Table 5.3). At birth, NBiW pigs
weighed 1.81 kg (SD 0.103) compared to 1.13 (SD 0.109) in LBiW pigs (P < 0.001).
By d 143 this difference had increased significantly, being almost an 8 kg difference
between LBiW and NBiW pigs. Low birth weight pigs exhibited significantly lower
ADG than NBiW pigs for all periods examined, with the exception of d 100 to 143,
where there was no difference in the performance of the two BiW categories (P > 0.05);
however the ADG of LBiW pigs was still slightly lower.
Milk supplementation (S or N) had no effect on the BW or ADG of LBiW or NBiW
piglets in MX litters, nor was there any interaction between BiW and milk
supplementation. Neither milk supplementation nor BiW had a significant effect on the
CV of BW for all periods examined. However there was a significant interaction
between BiW and milk supplementation for CV of BW, for all d examined with the
exception of d 1 (Table 5.3). From d 14 to 143 the CV of LBiW pigs with milk
supplementation was less than those without milk. In contrast, the CV of NBiW pigs
with milk supplementation was greater than those without milk.
65
Table 5.3 The effect of birth weight and milk supplementation on the performance of pigs in MX litters (litters with both low birth weight
(LBiW) and normal birth weight pigs (NBiW)) from d 1 to 143. Low birth weight pigs were ≤ 1.25 kg and NBiW pigs were between 1.60
and 2.0 kg. One half of the MX litters were given supplementary milk (S) during lactation and the other one half was not (N) 1
Birth
weight LBiW NBiW
Significance
Milk S N S N SEM Birth weight Milk Birth weight x Milk
BW, kg
Day 1 1.13 1.15 1.81 1.80 0.014 < 0.001 0.685 0.385
Day 14 3.64 3.89 5.01 5.15 0.198 < 0.001 0.323 0.561
Day 28 6.73 6.87 8.86 9.15 0.323 < 0.001 0.556 0.754
Day 49 14.0 14.4 17.6 17.9 0.533 < 0.001 0.305 0.812
Day 100 45.5 45.6 51.7 52.3 1.01 < 0.001 0.657 0.787
Day 143 72.4 73.7 80.6 80.6 1.53 < 0.001 0.737 0.736
ADG, kg/d
Day 1-14 0.193 0.211 0.247 0.258 0.015 0.002 0.294 0.789
Day 14-28 0.221 0.213 0.275 0.285 0.015 < 0.001 0.968 0.201
Day 28-49 0.348 0.358 0.414 0.419 0.021 0.004 0.612 0.765
Day 49-100 0.617 0.611 0.669 0.673 0.012 0.001 0.657 0.681
Day 100-143 0.624 0.653 0.672 0.659 0.032 0.411 0.764 0.553
CV
Day 1 9.92 10.1 5.67 4.97 1.51 0.004 0.745 0.768
Day 14 11.6 16.9 12.4 9.56 1.67 0.100 0.437 0.038
Day 28 13.8 18.6 15.2 8.81 2.09 0.051 0.475 0.011
Day 49 8.25 15.1 12.0 5.66 1.65 0.111 0.735 < 0.001
Day 100 6.21 11.5 9.65 5.26 1.10 0.432 0.547 < 0.001
Day 143 5.21 9.57 7.50 5.16 1.52 0.468 0.136 0.023 1Piglets were grouped in litters from d 1 until d 70; following this 2 litters selected at random were combined.
66
No difference was observed in the number of supplementary milk feeding episodes for
NBiW and LBiW pigs in MX litters (7.42 versus 7.86; P > 0.05). On d 27 there was an
effect of BiW on supplementary milk feeding duration, with LBiW piglets drinking for
longer than NBiW pigs (32.97 s, SD 3.10 versus 20.09 s, SD 2.89; P < 0.001).
A comparison of the performance of LBiW piglets in L litters with NBiW piglets in MX
litters, showed an identical pattern of results to those of LBiW and NBiW piglets in MX
litters. Birth weight has a significant effect on performance throughout with the
exception of d 100 to 143.
5.4 Discussion
This study investigated the effects of litter composition and milk supplementation
during the suckling period on the behaviour and growth performance of LBiW pigs to
weaning, and their long term effects to slaughter. It was hypothesized that there would
be an interaction between litter composition and milk supplementation, with LBiW pigs
in mixed litters being more likely to benefit from milk supplementation due to greater
competition from heavier littermates for limited resources (Algers and Jensen, 1991;
English, 1998; Lay et al., 2002). The results suggest that: (i) The WW of LBiW piglets
can be increased when grouped with similar weight littermates, although the effects do
not persist to slaughter and (ii) the provision of supplementary milk does not improve
performance but can reduce weight variation in LBiW pigs in mixed litters.
Consistent with the results of Milligan et al (2001b) and Kirkwood et al (2005), we
found no effect of litter composition on piglet mortality. Other studies have reported
reduced survival of LBiW piglets in mixed litters and differences in the weight
classification of LBiW piglets could be the cause. Our study defined LBiW piglets as ≤
1.25 kg, but previous studies have selected pigs of lower weights, for example < 1.0 kg
(Milligan et al., 2002a; Milligan et al., 2002b). As piglet BiW decreases this can have a
significant effect on performance (Paredes et al., 2012; Douglas et al., 2013) and
survivability (Bilkei and Biro, 1999). Therefore small differences in the BW of LBiW
pigs may affect mortality.
Low birth weight piglets benefitted from being cross fostered into a litter with other
LBiW piglets, specifically during the latter part of the preweaning period, exhibiting
67
higher ADG and greater WW. The results support the hypothesis of Cutler et al (1999)
and Fraser et al (1979 and 1995) that LBiW piglets are at a competitive disadvantage
when raised with heavier littermates. In contrast, LBiW pigs given access to
supplementary milk demonstrated no improvement in performance. This was contrary
to the expectation that LBiW pigs would benefit from additional milk, in particular
those in mixed litters. However, there was an interaction between BiW category and
milk supplementation on the CV of mixed litters. LBiW piglets in these litters with
supplementary milk had a lower CV from d 14 to 143 than those without supplementary
milk. Such a benefit was not observed in NBiW piglets in mixed litters which instead
saw an increase in CV. This treatment interaction suggests that supplementing mixed
litters with milk during lactation can decrease variation to slaughter in LBiW pigs,
which would be advantageous.
It is common farm strategy, in Europe at least, to cross foster piglets to create
littermates of similar weights, with the aim of reducing mortality and improving
performance. However research into this area has provided contradictory results. Whilst
Milligan et al (2001b) reported no statistical difference in the weight gain of LBiW
piglets whether grouped with heavier or similar sized littermates, there was a tendency
for LBiW piglets to gain more when grouped with heavier pigs. As this weight gain was
most prominent in smaller litters with only 8 or 9 piglets, maternal resources of the sow
were less likely to be limited and therefore piglets wouldn’t have been exposed to the
same level of competition as they would in a larger litter. In contrast with the results of
Milligan et al (2001b) this paper and others (Deen and Bilkei, 2004; English and Bilkei,
2004), found a decrease in the growth performance of LBiW piglets when grouped with
heavier litters. As effects were only observed in the latter part of lactation, it is unlikely
that direct competition from heavier littermates was the cause of poor performance of
LBiW piglets in MX litters. As ownership of a teat is usually established within the first
few days after birth (McBride, 1963) any effect of direct competition for access to a teat
would have been expected within the first week following parturition. Instead sow milk
is likely to be the limiting factor for piglet growth, with a plateau in the amount
available from d 21 onwards as piglet demand increases (Klobasa et al., 1987). As pigs
with heavier BiW can command higher quality teats, smaller piglets are more likely to
suckle from the less productive posterior teats (English et al., 1977; Fraser, 1984).
Additionally the BW of the pig may affect how well piglets can stimulate the teat.
Algers and Jensen (1984) proposed the ‘restaurant hypothesis’, in which the individual
68
piglet effectively orders the size of its next meal by massaging its own teat post ejection.
Therefore heavier piglets which are able to drain and massage the teats more vigorously
(Fraser, 1984), are more likely to stimulate milk production and have access to a great
amount of milk at their teat at the expense of lighter pigs (Drake et al., 2008), resulting
in an unequal distribution of milk across teats in mixed litters.
Additionally, it was observed that MX litters suckled less frequently in comparison to
litters composed of all LBiW piglets. The consequence of this is likely to be poorer
performance for LBiW piglets in these litters unless they consume more milk per
suckle (Campbell and Dunkin, 1982). The cause of this difference in suckling bout
frequency remains unclear. One possible explanation is that L litters initiate a higher
number of suckling bouts towards the end of lactation, as they require a greater amount
of milk than they are able to stimulate. Whilst in the first few days following parturition
suckling is initiated by the sow (Fraser, 1980), as the piglets’ age they are more likely to
attempt to instigate suckling although this is not always successful (Marchant-Forde,
2008). If LBiW piglets drink less per suckle (Campbell and Dunkin, 1982), they may
therefore be more likely to solicit additional suckling bouts from the sow. More
information is needed however to confirm the effect of piglet weight on suckling
frequency and the reasons behind this.
Despite an advantage in the WW of LBiW pigs grouped with other LBiW pigs, there
was no significant benefit observed for performance to slaughter. A potential
explanation for this lack of effect post weaning is that, although numerical differences
in BW are maintained to finishing, these relatively small differences cannot be detected
in the latter stages due to the increasing weight variation, as put forward by Wellock et
al (2009).
Nutrient intake of piglets can be limited during the lactation period which can have a
negative effect on growth performance (Pluske et al., 2005). Therefore providing
additional nutrition such as supplemental milk replacer can result in improved BW gains
to weaning (Azain et al., 1996; Zijlstra et al., 1996; Dunshea et al., 1998; Dunshea et al.,
1999; Wolter et al., 2002). However, whether any benefits persist in the long term
remains uncertain. In this study there was no effect of supplementary milk on the
growth performance of piglets irrespective of BiW or litter composition. Despite this, a
difference in the behaviour of LBiW piglets in different litter types was apparent, with
69
LBiW in MX litters drinking the supplementary milk for a longer period than both
heavier littermates and LBiW pigs in L litters. There are several possible explanations
for this change in behaviour, but absence of any benefit. First, it has been suggested
that, as a result of nutrient restriction in utero, LBiW pigs have a reduced capacity for
growth (Foxcroft et al., 2006). However, as an improvement was observed in the
performance of LBiW pigs in L litters this is unlikely. Second, it is possible that piglets
did not consume enough milk for any difference in ADG to be observed, especially as in
comparison to previous studies, milk intake was low (Azain et al., 1996; Wolter et al.,
2002; Miller et al., 2012). The ADMI for piglets in both L and MX litters during
lactation was 167 ml/d, whereas Azain et al (1996) had intakes of 471ml/d in litters
where the greatest effect on performance was observed. The provision of creep feed
during lactation may have had an effect on supplementary milk consumption. Limited
nutrient availability pre weaning is a major determinant of ADG during this period
(Klindt, 2003), however if piglets are receiving sufficient additional nutrition from
creep feed, this may reduce their supplementary milk intake. This is supported by the
fact that previous studies which saw a positive effect of supplementary milk on
performance did not provide creep feed (Azain et al., 1996; Zijlstra et al., 1996;
Dunshea et al., 1998; Dunshea et al., 1999).
Whilst the results presented here demonstrate that the performance of LBiW pigs can be
improved, in comparison to NBiW pigs they still had poorer growth rates and were
unable to catch up. Heavier BiW pigs not only retained their BW advantage from, birth
but this difference increased with age. This has been observed previously, with
preweaning management improving growth performance but inevitably BiW always
plays a greater role (Wolter et al., 2002), with pigs that are inherently heavier
performing better (Lawlor et al., 2002). Only in latter part of the finishers a similar
growth rate was observed between the two BiW groups, an observation which has been
noted in previous studies (Gondret et al., 2005). One possibility is that the pigs’
environment in the later stages imposes a constraint and differentially affects pigs of
varying weights. For example, pigs of greater BW have a lower thermoneutral zone
(Baker, 2004) and, in situations of high environmental temperature, will show a greater
drop in FI and thus a reduced growth performance (Nienaber et al., 1996).
70
5.5 Conclusions
This paper offers novel insights in the management of LBiW pigs and their behaviour
during the lactation period when provided with supplementary milk. First, the results
suggest that the WW of LBiW pigs can be increased by cross fostering LBiW piglets
with similar weight littermates, however this BW advantage does not persist long term.
Second, the addition of supplementary milk does not benefit their growth, but does
reduce the variation in BW of pigs and an increase in the duration of supplementary
milk intake was noted for LBiW pigs in mixed litters.
71
Chapter 6: High specification starter diets improve the performance of
low birth weight pigs to 10 weeks of age
6.1 Introduction
Starter regimes are critical to minimize any growth check at weaning, as well as to
maximize FI and growth performance not only in the nursery period but also to
slaughter (Lawlor et al., 2002). In the UK, a starter regime typically consists of a series
of diets fed in succession formulated for the ‘average pig’, which decrease in cost and
specification. The aim of such regimes is to allow pigs to reach a particular BW after
which they can progress to a cheaper compound diet, all in a relatively short period of
time.
Low birth weight pigs are likely to be substantially lighter at weaning than NBiW pigs
(Douglas et al., 2013; Douglas et al., 2014). As a result of their reduced BiW, LBiW
pigs may have immature digestive systems, including reduced secretion of digestive
enzymes at birth (Xu et al., 1994; Wang et al., 2008; D'Inca et al., 2010b) and a less
developed digestive tract at weaning (Michiels et al., 2013) hindering adaptation to post
weaning diets compared to heavier pen mates (Pluske et al., 2003). Recent research
suggests LBiW pigs can improve growth performance when provided with a better
quality regime (Beaulieu et al., 2012); however whether these improved regimes are
economical is uncertain. Similarly, high allowances of starter regimes in comparison to
low allowances have positive effects on performance of light pigs from weaning
(Lawlor et al., 2002; Magowan et al., 2011b) although the effect on exclusively LBiW
pigs is unknown.
The aim was therefore to investigate whether a high specification starter regime and the
provision of an extra amount of feed (corresponding to the last feed in the regime) can
have additive or synergistic effects on the performance of LBiW pigs at weaning. The
cost of the different feeding regimes was investigated to determine if any successful
treatments were economically viable. It was hypothesized that LBiW pigs would benefit
most from both a high specification regime as well as an increased allowance of the
final diet in the starter regime.
72
6.2 Materials and method
6.2.1 Experimental design
The experiment was conducted at Cockle Park Farm, Newcastle University and was
approved by the Animal Welfare and Ethics Review Board at the University. The
experiment was designed as an incomplete 2 x 2 x 2 factorial with 6 replicates; a total of
180 crossbred pigs (dam was Large White x Landrace cross and sire was Hylean
synthetic, Hermitage Seaborough Ltd., Devon, UK) were used. The factors used were
starter regime (High specification starter regime [H] or Standard starter regime [S]),
birth weight (Low birth weight [BWL] or normal birth weight [BWN]) and increased
allowance of the final diet in the starter regimes (Feed 3) (Yes [Y] or no [N]). None of
the BWN pigs received the increased allowance of feed 3 resulting in an incomplete
experimental design.
6.2.2 Animal management
Sows were farrowed on a three week cycle in individual farrowing crates which were
equipped with a feeder and drinker for the sow. The nutrition of the sows was
standardized across all treatments, with identical diets fed before and during lactation.
The temperature in the farrowing house was maintained at 21°C by a centrally
controlled heating and ventilation system; an infra-red heat lamp was located in the
creep area for the piglets to provide a microclimate during the lactation period.
Efforts were made during the lactation period to maximize the growth of all pigs by
reducing limiting factors such as competition from heavier littermates and poor milk
supply. A feeder containing supplementary liquid milk for the piglets was added to the
pen of all litters from 24 hours after birth and was removed on d 7. Piglets were
observed for signs of diarrhoea (nutritional scour) and other illness related to the milk
supplement. Any piglets that exhibited diarrhoea were treated with 0.5 to 1 ml,
depending on BW, of Norodine (Norodine 24 solution for injection, Norbrook, Corby,
UK): if 3 or more piglets in a litter showed symptoms, the whole litter was treated.
Within the first 12 h of birth, pigs were teeth clipped, weighed and ear tagged with
metal chick wing tags for identification. Male pigs were not castrated. Morphometric
measurements were also taken: CRL, snout-ears length and abdominal and cranial
circumference. Piglets were tail docked and given an iron injection at d 3. From d 10
onwards, a small amount of creep feed was placed on the floor of the heated creep area
73
once a day and this was an equal mix of the first diet of starter regimes H and S (see
below for details).
6.2.3 Experimental management
Approximately 17 sows farrowed per batch and piglets from these sows were weighed
within 12 h of birth. Low birth weight piglets were classified as ≤ 1.25 kg (the minimum
BW was 700 g) and normal birth weight piglets as 1.6 to 2.0 kg (Douglas et al., 2013);
piglets that did not meet the weight criteria were cross fostered onto non experimental
sows. According to the normal husbandry practice on farm, during the first 24 h after
birth all piglets selected for trial were cross fostered into one of three litters according to
birth weight, so experimental litters contained either low BWL or BWN piglets. Litter
size was set at 11 or 12 piglets, depending on the number available.
Pigs were weaned at d 28 (+/- 1 d) of age, when they were transferred to controlled
environment nursery accommodation with slatted plastic floors; each pen provided 0.42
m2/pig and had a feeder with 3 spaces and a separate nipple drinker giving ad libitum
access to water. If pigs did not reach 4 kg by d 28 then they were not weaned and
removed from trial (5 BWL and 2 BWN in total). Ear tags were removed and replaced
with plastic weaner tags and pigs were vaccinated for Mycoplasma hyopneumoniae and
Porcine Circovirus Type 2 (Inglevac Mycoflex, Boehringer Ingelheim, Germany). A
thermostatically controlled heating and fan ventilation system initially maintained the
room temperature at 26 °C; this was then reduced by 0.2 °C/day to 22 °C.
On d 28 piglets, when weaned and moved to experimental accommodation, pigs were
randomly assigned within birth weight category to form treatment groups of 5 pigs per
pen balanced for sex and litter of origin using SAS Proc plan of version 9.2 (SAS
Institute Inc., Cary, NC) to randomly allocate pigs to treatments. Any additional pigs
were removed from the experiment. Pigs assigned to regime H or S received the
appropriate starter regime which was fed on a kg/pig basis; both BWL and BWN pigs
were assigned to these regimes. Feed was available to pigs at all times during the
experiment.
Pigs offered regime H were given 2.5 kg/pig of feed 1 (24.0 % CP, 17.3 MJ DE, 1.75 %
total lysine and 20.0 % lactose), 2 kg/pig of feed 2 (23.8% CP, 16.0 MJ DE, 1.60 %
74
total lysine and 15.0 % lactose) and 3 kg/pig of feed 3 (23.4 % CP, 15.3 MJ DE, 1.50 %
total lysine and 5.00 % lactose) (Primary Diets, Ripon, North Yorkshire) (Table 6.1 and
6.4). Pigs offered regime S were given 2 kg/pig of feed 2 (23.8% CP, 16.0 MJ DE, 1.60
% total lysine and 15.0 % lactose) and 3 kg/pig of feed 3 (23.4 % CP, 15.3 MJ DE, 1.50
% total lysine and 5.00 % lactose). Feeds were offered in succession, with the next feed
provided once the previous had been consumed.
75
Table 6.1 Ingredient composition on an as-fed basis and chemical analysis of the four
diets used1
Diet
Item 1 2 3 W
Ingredient, g/kg
Micronized barley 75.0 75.0 75.0 150.0
Wheat - 234.1 438.1 487.6
Micronized wheat 150.0 50.0 25.0 -
Micronized maize 25.0 25.0 - -
Porridge oats 100.0 75.0 25.0 -
Wheat feed - - 12.5 25.0
Herring meal 75.0 75.0 60.0 25.0
Hi-pro soya 94.4 145.2 223.3 250.0
Full fat soya bean 65.6 25.0 25.0 -
Pig weaner vitamin/trace element supplement2 5.00 5.00 5.00 5.00
Whey protein concentrate 50.0 - - -
Dried Skim Milk Powder 75.0 61.3 - -
Whey 195.8 173.2 69.4 -
Potato protein 12.5 12.5 - -
L-lysine 2.30 1.68 2.45 3.74
DL-Methionine 2.11 1.45 1.31 1.56
L- Threonine 1.53 1.15 1.19 1.57
L-Tryptophan 0.42 0.23 0.01 0.13
Vitamin E 0.41 0.41 0.21 0.10
Benzoic acid 5.00 5.00 5.00 5.00
Limestone flour - 0.76 - 1.09
Dicalcium phosphate - - 5.11 8.94
Salt - - 1.21 4.06
Binder (lignobond DD) - - - 6.25
Soya oil 65.0 33.2 25.2 25.0
Analysed composition, % as fed3
CP 22.4 21.7 21.0 18.0
Crude Fibre 1.80 2.00 2.60 2.90
Moisture 9.60 11.0 11.7 11.9
Ash 5.60 5.50 5.10 4.70
Calculated composition, % as fed or as specified4
DE, MJ/kg 17.30 16.00 15.30 14.80
Calcium 0.62 0.59 0.54 0.59
Phosphorous 0.60 0.59 0.54 0.59
Lactose 20.00 15.00 5.00 0.00
Lysine 1.75 1.60 1.50 1.40
Methionine 0.67 0.60 0.54 0.50 1Diets were supplied by Primary Diets, Ripon, North Yorkshire
2 It provided per kg of complete diet: 11,500 IU of vitamin A, 2,000 IU of vitamin D3,
100 IU of vitamin E, 4 mg of vitamin K, 27.5 µg of vitamin B12, 15 mg of pantothenic
acid, 25 mg of nicotinic acid, 150 µg of biotin, 1.0 mg of folic acid, 160 mg of CU
(CUSO4), 1.0 mg of Iodine (KI, Ca(IO3)2), 150 mg of Fe (FeSO4), 40 mg of Mn (MnO),
0.25 mg of Se (BMP-Se), 110 mg Zn (ZnSO4) 3Proximal analysis performed by Sciantec Analytical Services Ltd (North Yorkshire)
4Values estimated using raw material matrix (Primary Diets, Ripon, North Yorkshire)
76
Once BWL pigs had consumed their starter regime allocation they were either fed an
extra 2.5 kg of feed 3 (treatment Y) or not (treatment N). The consequence of the
incomplete experimental design was that whilst BWL pigs were assigned to both
regimes, BWN were not given access to the extra feed 3. By d 49 pigs had consumed all
experimental diets, irrespective of treatment. Once the allocated starter feeds had been
consumed, pigs were given a weaner feed (W; 21.6 % CP, 14.8 MJ DE and 1.40 % total
lysine) which was fed ad libitum until d 70. Thus there was a logical progression in the
composition of the feeds 1 to W, with the former being based on more high quality
ingredients and greater nutrient concentration. Table 6.1 reports the composition and
chemical analysis of all diets used for each feeding regime. The chemical analysis of the
feeds was performed in accordance with AOAC (AOAC, 1990) standards using
proximate analysis and the calculated composition was estimated from the values in the
Premier Atlas ingredients matrix (Hazzledine, 2008).
From d 1 to 70, pigs were individually weighed twice a wk, on a Monday and Thursday
morning. Feed intake per pen was measured twice a week from d 28 to 70 by weighing
residual feed at the same time that pigs were weighed. With these measurements, ADG
was calculated for both individual animals and treatment groups, and ADFI for pen
groups only. The CV of BW was calculated for within pen (5 piglets per pen).
Individual morphometric measurements taken at birth were taken to calculate the
relative CRL (CRL/BW) and ponderal index (PI) (BW/CR3). The period d 28 to 49 and
49 to 70 were chosen for statistical analyses as by d 49 all pigs, regardless of treatment,
had consumed their experimental diets; therefore from d 49 to 70 is when animals were
only consuming feed W.
For the economic analysis, the cost per tonne of each feed was provided by the company
that manufactured the feed (based on the raw material costs, correct as of November
2013); this was then used to calculate the feed cost per pig of each regime. The cost per
kg of BW gained was then calculated as: feed cost per pig / BW gain per pig. Finally the
margin over feed (MOF) was calculated by the following formula: MOF = [BW gain x
proportion of carcass weight from live weight (0.75) x current pig price per kg carcass
weight (as of Nov 2013, £1.70) - feed cost per pig].
77
6.2.4 Statistical analysis
All performance data was tested for normality using the Univariate procedure of SAS
version 9.2 (SAS Institute Inc., Cary, NC) and was normally distributed. Sex was
included as a factor in all preliminary models was not significant so omitted for
subsequent analysis. Data for all analyses were blocked by farrowing batch (6 batches).
Treatment x batch interactions were added to all preliminary models, but were not
significant and therefore omitted from subsequent analysis. Differences were considered
significant at P < 0.05 and reported as a tendency towards statistical significance at P <
0.10. Data are presented as least square means.
The experimental design addressed two hypotheses. The effects of the different feeding
regimes and their interactions, for BWL pigs only, were analysed as a repeated measure
ANOVA using Proc Mixed of SAS version 9.2 (SAS Institute Inc., Cary, NC). Starter
regime (H or S), extra feed 3 (Y or N) and time were added as factors. Two separate
models were run with data analysed on a litter basis from d 1 to 28 (11 or 12 piglets)
and on a pen basis from d 28 to 70 (5 pigs). To investigate whether BiW differentially
affected H or S pigs, a repeated measures ANOVA using Proc Mixed of SAS was used
with BiW category (BWL or BWN) and starter regime (H or S) as factors using the
same methods as previously described.
To test whether any of the feeding regimes allowed BWL pigs to catch up with BWN
pigs, a comparison between the best regime (regime HY) and BWN pigs on either
regime H or S was made by using a set of orthogonal contrast statements.
An economic analysis was run as a one way ANOVA to allow comparison of all
treatment combinations (that is starter regime, extra feed 3 and BW). All statistical
analyses were conducted at a significance level of 5%. Data are presented as LS means
and differences were considered significant at < 0.05 and reported as a tendency
towards statistical significance at < 0.10.
6.3 Results
Total pre weaning mortality after litter allocation was 3.4% and similar between BWL
and BWN piglets (3 versus 4 pigs died). No pigs were removed from the trial due to
illness or other causes from d 28 to d 70. At birth, BWL pigs weighed 1.08 kg (SD
78
0.150) versus 1.82 kg (SD 0.098) in BWN pigs (P < 0.001). At weaning BWN were
almost 1.5 kg heavier than BWL pigs, with a strong correlation between BW at birth
and weaning (r = 0.695; P < 0.001) and those pigs lighter at birth weighing significantly
less at weaning (7.15 kg, SD 1.29 vs. 8.62 kg, SD 1.23; P < 0.001)). As well as
differences in BW at birth, BWL pigs had a shorter CRL of 24.2 cm (SD 1.50)
compared to 27.2 cm (SD 1.67) for N piglets (P < 0.001). However BWL pigs had a
significantly greater CRL/BW, (21.6 cm/kg, SD 2.01) than BWN pigs (15.2 cm/kg, SD
1.34), and a significantly lower PI (81.1 kg/m3, SD 8.3 vs. 91.4 kg/m
3, SD 9.5) (P <
0.001). The mean age of pigs at weaning was 28 d (± 1 d) and at the end of the nursery
period 70 d (± 1 d).
6.3.1 Performance of low birth weight pigs
Starter regime had a significant effect on the ADG of BWL pigs between d 28 and 49
(Table 6.2), with pigs that received the starter regime H performing better than those
that did not (regime S) (ADG = 0.397 vs. 0.362 kg/d respectively). The only residual
effect seen on the performance of the pigs between d 49 to 70 was due to starter regime
on ADG (H = 603 kg/d, S = 664 kg/d; P = 0.017). From d 28 to d 49, the provision of
the extra feed 3 also resulted in improved performance, with pigs fed the extra feed 3
exhibiting a significantly greater ADG (0.399 kg/d) than those which were not (0.360
kg/d). There was an interaction between starter regime and extra feed 3 (P = 0.029)
from d 28 to 49). The interaction was due to the fact that pigs fed starter regime H with
(treatment HY) exhibited the greatest daily gain (0.452 kg) in comparison to all other
treatments (P = 0.029), whilst S pigs showed much less response to additional feed 3.
In contrast, there was no effect of starter regime or additional feed 3 on the ADFI of
BWL pigs from d 28 to 70. However there was an interaction, with pigs on regime HY
and SN eating more than those on HN or SY (P = 0.026). Whilst there was no effect of
starter regime or an interaction during this period on FCE, there was a significant effect
of extra feed 3 provision, with those that received extra showing an improved FCE in
comparison to those which did not (Y = 0.93, N = 0.83; P < 0.030).
Whilst there was no effect of starter regime on BW at d 49, by d 70, there was a
significant effect with pigs fed starter regime H weighing almost 2 kg more than those
on S (29.0 kg versus 27.3; P = 0.014). In contrast, the effect of provision of extra feed 3
on BW tended to be significant at d 49 (Y = 15.27 kg, N = 14.67 kg; P = 0.094) and was
79
significant at d 70, with and those given extra feed 3 weighing over 1.5 kg heavier than
those pigs which were not at the end of (28.9 kg versus 27.4; P < 0.027). There was no
effect of starter regime, extra feed 3 or an interaction, on the CV of BW of BWL pigs
from d 29 to 70.
80
Table 6.2 The effect of starter regime and additional allowance on the growth performance of low birth weight pigs, from d 1 to d 70 1, 2, 3
Treatment Significance
Item H Y H N S Y S N SEM SR AF SR x AF
BW, kg
D 1 1.09 1.10 1.06 1.05 0.032 0.172 0.911 0.817
D 28 6.85 7.14 7.26 7.15 0.177 0.241 0.613 0.271
D 49 15.8 14.7 14.7 14.6 0.340 0.103 0.094 0.145
D 70 30.1 27.9 27.7 26.9 0.629 0.014 0.027 0.320
ADG, kg/d
D 1 to 28 0.213 0.224 0.230 0.226 0.007 0.177 0.642 0.310
D 28 to 49 0.452 0.361 0.365 0.358 0.014 0.019 0.010 0.029
D 49 to 70 0.688 0.640 0.619 0.586 0.024 0.017 0.100 0.748
D 28 to 70 0.560 0.500 0.492 0.472 0.013 0.001 0.005 0.129
ADFI, g/d
D 28 to 49 452 418 410 447 21.4 0.770 0.948 0.110
D 49 to 70 1,049 967 998 1,011 39.8 0.834 0.467 0.203
D 28 to 70 742 679 673 713 21.4 0.427 0.610 0.026
FCE
D 28 to 49 0.970 0.871 0.897 0.801 0.044 0.102 0.030 0.969
D 49 to 70 0.658 0.674 0.634 0.584 0.041 0.196 0.780 0.430
D 28 to 70 0.756 0.743 0.737 0.665 0.032 0.149 0.209 0.376
CV
D 1 11.7 10.3 13.4 11.5 2.19 0.385 0.346 0.895
D 28 18.7 13.9 19.3 17.6 2.71 0.438 0.243 0.577
D 49 12.9 11.7 12.7 14.3 1.28 0.369 0.879 0.282
D 70 10.2 10.4 11.0 13.1 1.12 0.133 0.320 0.433 1 Starter regime: H (high specification regime) = 2.5 kg feed 1, 2.0 kg feed 2, 3.0 kg feed 3; S (standard commercial regime) = 2.0 kg feed
2, 3.0 kg feed 3 2 Additional feed 3: N = no additional feed 3, Y= Yes additional feed, 2.5 kg feed 3
3 Low birth weight (BWL) pigs were selected at birth (≤ 1.25 kg)
81
6.3.2 Comparison of low birth weight and normal birth weight pigs
There was a significant effect of BW category on ADG during all periods examined,
with BWN performing better; although during the period d 49 to 70 this was only a
tendency (BWL = 0.613 kg/d, BWN = 0.659 kg/d; P = 0.07). Similarly, there was an
effect of BiW category on the BW of pigs from d 1 to 70, with BWL pigs weighing less
at all periods examined (P < 0.001) (Table 6.3). At birth, BWN piglets weighed 1.82 kg
(SD 0.095) compared to 1.08 kg (SD 0.150) for BWL piglets. By d 70 this difference
had increased to over 3 kg in BW (P < 0.001). Whilst BWL pigs ate less than BWN
pigs for all periods examined (P < 0.05), there was no effect of body weight category on
FCE (P > 0.05).
There was no effect of starter regime or interaction between starter regime and birth
weight category of pigs on any of the performance parameters measured (ADG, ADFI
or FCE; P > 0.05) from d 28 to 70. However there was also a tendency for an
interaction between BW category and starter regime for ADG from d 49 to 70 (P =
0.095), with BWL pigs on regime H and BWN pigs on regime S performing better than
pigs on other treatments.
Orthogonal contrasts comparing the best performing BWL pigs (treatment HY) and the
combined effect of BWN pigs (treatments H and S), showed that by d 49 there was no
significant difference in BW (15.8 kg versus 16.9; P = 0.135). By d 70, BWL pigs on
treatment HY had completely caught up with normal weight pigs (H and S) (30.0 kg
versus 30.6; P = 0.413). In addition, BWL on regime HY consumed a similar total
amount of feed as BWN pigs (H and S) from d 28 to 49 (452 g/d versus 515; P =
0.118), but less during d 49 to 70 (1049 g/d versus 1174; P = 0.040). For both periods
examined, BWL pigs on regime HY had an improved FCE in comparison to BWN pigs
(d 28 to 49 = 0.97 versus 0.80; P < 0.001, d 49 to 70 = 0.66 versus 0.57; P = 0.045),
even though consuming the same diet during the latter period.
82
Table 6.3 The effect of starter regime on the performance of low birth weight and normal birth weight pigs at weaning, from d 1 to d 70 1, 2
Treatment Significance
Item BWL H BWL S BWN H BWN S SEM BW SR BW x SR
BW, kg
D 1 1.10 1.05 1.81 1.82 0.035 < 0.001 0.584 0.385
D 28 7.14 7.15 8.57 8.67 0.170 < 0.001 0.306 0.331
D 49 14.7 14.6 16.9 16.8 0.498 < 0.001 0.855 0.958
D 70 27.9 26.9 30.3 30.9 0.875 < 0.001 0.809 0.347
ADG, kg/d
D 1 to 28 0.221 0.226 0.235 0.253 0.008 0.021 0.189 0.449
D 28 to 49 0.361 0.358 0.415 0.396 0.019 0.025 0.582 0.679
D 49 to 70 0.640 0.586 0.644 0.674 0.024 0.066 0.608 0.095
D 28 to 70 0.496 0.467 0.530 0.535 0.018 0.012 0.540 0.375
ADFI, g/d
D 28 to 49 418 447 512 518 29.5 0.012 0.554 0.695
D 49 to 70 967 1,011 1,157 1,190 49.7 0.001 0.446 0.916
D 28 to 70 693 729 834 854 31.9 < 0.001 0.387 0.792
FCE
D 28 to 49 0.871 0.801 0.818 0.773 0.035 0.266 0.118 0.730
D 49 to 70 0.674 0.584 0.560 0.571 0.039 0.105 0.307 0.191
D 28 to 70 0.743 0.665 0.637 0.629 0.030 0.095 0.148 0.230
CV
D 1 9.31 11.5 4.89 5.67 1.69 0.007 0.385 0.670
D 28 13.9 17.6 14.3 9.71 3.17 0.249 0.892 0.209
D 49 11.7 14.3 11.5 9.51 1.40 0.089 0.826 0.119
D70 10.4 13.1 10.5 8.83 0.983 0.045 0.610 0.042 1 Low birth weight (BWL) and normal birth weight (BWN) pigs were selected at birth (BWL = ≤ 1.25 kg, BWN = 1.6 to 2.0 kg)
2 Starter regime: H (high specification regime) = 2.5 kg feed 1, 2.0 kg feed 2, 3.0 kg feed 3; S (standard commercial regime) = 2.0 kg feed
2, 3.0 kg feed 3
83
6.3.3 Economic analysis
The most expensive starter regime was regime H which had a total feed cost of £6.15
per pig in comparison to regime S which was the cheapest at £3.48 per pig (Table 6.4).
The cost of the additional feed 3 was £1.47 per pig. This meant that the most expensive
treatment combination was treatment HY, which cost £7.62 in total. However when
considering the amount of feed W consumed post treatment, BWN pigs on regime H
were the most expensive (£15.9), whilst BWL pigs fed regime SY had the lowest total
feed costs at £12.7 per pig. There was a tendency towards significance (P = 0.087) for
the cost per kg of BW gained, with BWL on regime SY having the lowest in
comparison to BWN pigs fed regime H which was the highest (0.62 £/kg vs. 0.72 £/kg).
Whilst there was no statistical difference between the treatments for the margin over
feed cost (P > 0.05), the highest value was for BWL pigs fed regime HY, whilst the
lowest was for BWL pigs fed regime SN (£14.5 versus 12.3).
84
Table 6.4 The effect of birth weight, starter regime and additional allowance on the mean cost of feeding a pig from d 28 to 70 and margin
over feed 1,2,3
BWL BWN Significance
Item HY H N SY SN H S SEM Treatment
Regime4
Feed 1 2.5 2.5 - - 2.5 - - -
Feed 2 2.0 2.0 2.0 2.0 2.0 2.0 - -
Feed 3 5.5 3.0 5.5 3.0 3.0 3.0 - -
Total starter (kg/pig) 10.0 7.50 7.50 5.00 7.50 5.00 - -
Total feed (kg/pig) 30.4 28.7 28.9 31.0 34.6 35.4 0.652 0.001
Starter diet costs, £/pig5 7.62 6.15 4.95 3.48 6.15 3.48 0.409 < 0.001
Total cost of feed, £/pig6 15.0 13.8 12.7 12.9 15.9 14.4 0.417 < 0.001
£/kg gain 0.65 0.66 0.62 0.65 0.72 0.65 0.096 0.087
Margin over feed7 £/pig 14.5 12.7 13.5 12.3 12.4 14.1 1.10 0.293
1 Low birth weight (BWL) and normal birth weight (BWN) pigs at weaning (BWL = ≤ 1.25 kg, BWN = 1.6 to 2.0 kg)
2 Starter regime: H (high specification regime) = 2.5 kg feed 1, 2.0 kg feed 2, 3.0 kg feed 3; S (standard commercial regime) = 2.0 kg feed
2, 3.0 kg feed 3. After which a standard feed W was offered ad lib in both regimes until d 70 3 Additional allowance: N = no additional feed 3, Y= Yes additional feed, 2.5 kg feed 3
4 Diets were fed in succession once the previous feed had been consumed, i.e. feed 1, feed 2 then feed 3
5 Diets 1, 2 and 3 cost per tonne: Feed 1= £1069, Feed 2 = £858, feed 3 = £588, Feed W = £360
6Pig prices (Nov 13): carcass price= £1.70/kg deadweight
85
6.4 Discussion
This study investigated the short to medium term effects of feeding a high specification
starter regime with or without extra feed on the performance from weaning of LBiW
pigs. Treatments were fed on a kg/pig basis, lasting approximately three weeks, to
assess if dietary manipulation could improve the nursery exit weight of these LBiW
pigs. Our hypothesis was that the combination of an improved starter regime and the
provision of the extra feed (corresponding to the last feed of the starter regime) would
improve the performance of LBiW pigs post weaning compared to a standard regime. It
is further hypothesized that this improved regime is unlikely to improve NBiW pig
performance or be economic. The hypothesis was based on the findings : 1) diets based
on cooked cereals, containing a higher lysine content and lactose appear to confer
advantages on LBiW pigs and 2) increased amounts of standard starter regimes have
positive effects on performance of light pigs from weaning (Lawlor et al., 2002;
Magowan et al., 2011b). The increase in performance of LBiW pigs fed an improved
starter diet was consistent with our hypotheses. Most importantly, the findings
demonstrate that the additional cost of feeding LBiW pigs improved diets is offset by
BW gains and is therefore economically viable.
Weaning weight is likely to be a reflection of both pre and post-partum factors (Lawlor
et al., 2002), with uterine environment as well as lactation management affecting
preweaning growth performance. Selecting piglets by birth weight (low (≤ 1.25 kg) or
normal (1.6 to 2.0 kg)) for treatment at weaning, meant that pigs were more likely to
have a similar underlying cause for low weight at weaning (birth weight) and therefore
similarities in their nutritional requirements, rather than selecting pigs based solely on
WW which may be due to a number of reasons. An effort was made to reduce limiting
factors during the preweaning stage to give LBiW pigs every chance to grow to their
full potential. Piglets were cross fostered into litters by BiW to reduce the competition
from heavier littermates and supplementary milk was also provided. However as
expected, BiW was a good determinant of WW (Gondret et al., 2005; Bérard et al.,
2010; Douglas et al., 2013), with a high correlation between the two. At birth, there was
an average difference of 0.7 kg between the two weight categories, however by d 28 this
had increased to almost 1.5 kg. Whilst the majority of pigs born with low BW remained
light at weaning, there was a small minority of pigs which met or exceeded the average
BW of NBiW pigs at weaning, possibly as a result of the experimental design which
reduced limiting factors.
86
Light weight pigs can be problematic at any stage of production, and the question of
how best to feed these pigs at weaning and beyond to ensure maximum performance
often results in conflicting advice. It has been suggested, in accordance with the
predictive adaptive response hypothesis (Gluckman et al., 2005; Rehfeldt et al., 2012),
that as a result of restriction in utero LBiW pigs have reduced absolute nutrient
requirements, for example reduced dietary protein (Nissen and Oksbjerg, 2011). In
contrast, this study and another (Beaulieu et al., 2012) hypothesized that feeding LBiW
pigs a diet higher in nutrient concentration can actually benefit the performance of these
pigs at weaning. Whilst other studies have also investigated the effect of high input
dietary regimes (Magowan et al., 2011b) or high density diets (Lawlor et al., 2002) on
the performance of light pigs at weaning, these were not exclusively LBiW but pigs
which were light at weaning possibly as a result of other factors. Low weaning weight
influences digestive enzyme activity and gut maturation (Owsley et al., 1986; Mahan
and Lepine, 1991) with lighter pigs at weaning as a result of LBiW having a delayed gut
maturation (Michiels et al., 2013) which may hinder adaptation to post weaning diets,
and therefore likely to benefit from a highly digestible diet with high nutrient density.
In agreement with the above hypothesis, LBiW pigs at weaning benefitted significantly
from provision of a high specification feed. Identification of any individual ingredient
which may have benefitted BWL pigs is not possible; however high digestibility, high
lactose content and inclusion of cooked cereals have all been shown to improve nursery
pig growth performance (Mahan, 1992, 1993; Medel et al., 2004; Kim et al., 2010;
Menoyo et al., 2011), in particular for those pigs with immature guts as observed in
LBiW pigs at weaning (Michiels et al., 2013). Whilst it has been previously
demonstrated that high specification starter diets (Beaulieu et al., 2012) can improve the
performance of LBiW pigs at weaning, and increased amounts of standard starter
regimes (Lawlor et al., 2002; Magowan et al., 2011b, a) can improve the performance of
light weight pigs at weaning, a combination of both improved starter diet quality and
increased amount had not previously been investigated.
The results presented demonstrate that provision of a high nutrient dense and highly
digestible diet can improve performance of LBiW pigs, but this alone was not enough to
maximize growth of these pigs, and an extra provision of feed 3 was needed. Provision
of both of these nutritional treatments resulted in a 3 kg weight increment at nursery exit
and reduced variability in BW in LBiW pigs, and subsequently there was no significant
87
difference in BW compared to pigs which were heavier at birth and weaning. The
combined effects of the treatments on growth performance are likely to be the result of a
synergistic effect, as indicated by the interaction of starter regime and additional feed 3
on ADG, rather than simple additivity. The positive impact on performance is not
likely to be a result of an improved FI, as there was no difference in absolute
consumption between LBiW pigs on the different dietary treatments, but rather due to
an improved FCE as noted previously (Lawlor et al., 2002; Magowan et al., 2011b).
However as the high density does show improved performance, it could indicate that FI
is a limiting factor in LBiW pig performance as suggested by others (Gondret et al.,
2005).
Whilst an improvement in performance was observed for LBiW pigs on different
dietary regimes, there was no difference in growth performance of NBiW pigs fed
dietary regime H or S (without extra feed 3). This suggests that the ‘standard’ feeding
regime was appropriate for the NBiW pigs. It is apparent that pigs of different weight
respond differently to starter regimes of different specifications. Despite no differences
in the growth performance of BWN pigs on different dietary treatments, they still had
greater growth rates than BWL pigs on comparable treatments. This is in agreement
with previous work which has shown that BiW has a greater effect on post weaning
performance than a starter regime alone (Mahan et al., 1998; Lawlor et al., 2002). The
exception to this was the first two weeks post weaning where no difference in growth
performance was noted, as well as no difference in the FI during the first week post
weaning. It has been noted that heavier pigs at weaning often experience a greater
growth check than lighter pigs (Lewis and Wamnes, 2006)(although this was observed
in pigs weaned at d 17), possibly as a result of decreased consumption of creep feed as
they have access to more productive teats. Therefore they may not transition to solid
feed as well which results in a temporary drop in performance. However the design of
this experiment means this is unlikely to occur as pigs of different BiW categories were
in different litters so NBiW pigs will have been competing with similar sized littermates
for teat access.
A weight increment of over 3 kg for BWL pigs was observed at nursery exit as a result
of the nutritional treatments. However, subsequent performance to slaughter was not
investigated and therefore whether any gain in BW is retained remains uncertain.
Previously, we have demonstrated that pigs retain any increase in BW as a result of pre
88
weaning treatments (Douglas et al., 2014), although treatment differences were not
significant at slaughter, possibly as a result of the increased variation in BW at a greater
age (Wellock et al., 2009). In contrast, other studies have found that whilst pigs fed
starter diets which were higher in quality did benefit nursery performance, beyond this,
effects on BW and ADG diminished with age (Meade et al., 1969; Whang et al., 2000).
However, the improved growth of H pigs during the subsequent period on a standard
diet suggests longer term benefits may persist.
To be an economically viable treatment, the price of feeding the higher quality diets
must be offset by the gains in BW or improved food utilization (Lawlor et al., 2002).
Feeding LBiW pigs the high specification starter regime with extra feed 3 was the most
expensive; however the margin over feed was better in comparison to other regimes.
These results suggest that not only can LBiW pigs at weaning benefit from an improved
dietary regime, but that it is cost effective for producers with an increased return per
pig, and should be preferred to a standard commercial regime which has the poorest
margin over feed. In contrast, for normal weight pigs the standard commercial regime
was the least expensive, had the greatest margin over feed and resulted in the lowest
variability in BW.
6.5 Conclusions
Pigs of different BiW are better suited to different starter diet regimes and should be fed
as such. Based on the conclusions of both the growth performance and economic
analysis, the following management techniques can be recommended for improving the
growth performance of pigs with low weights at weaning as a result of LBiW: (1)
Feeding a high specification starter diet with extra feed 3 can not only result in a similar
nursery exit weight of low BW and normal BW pigs and reduced variability in BW, but
is the most economical diet regime, (2) Separation of pigs with low BW at weaning will
allow selective feeding of an improved regime as heavier pigs are best suited to a
standard commercial diet. Limiting the usage of the most expensive regime to low BW
pigs only, will ensure maximum growth performance as well as increased profitability.
89
Chapter 7. General Discussion
The aim of this thesis was to provide understanding of risk factors that are associated
with the occurrence of light weight pigs and, by providing such understanding, to
develop nutritional and management treatments that might enable these pigs to decrease
the deficit in their BW.
Initially the risk factors associated with poor lifetime performance were investigated
(Chapter 3). In particular, low BiW and WW resulted in poor growth to finishing, which
highlighted the need to focus on earlier stages of production. The effect of a high
specification diet (higher in amino acid: energy content) introduced at 9 weeks of age on
the subsequent growth of LBiW pigs was investigated, to determine if there is a critical
window in early life for interventions to improve growth performance (Chapter 4). As
LBiW pigs did not benefit from this nutritional intervention during the grower stage, the
effect of the preweaning environment on performance to slaughter was explored
(Chapter 5). Whilst reducing competition from heavier littermates increased the WW of
LBiW pigs, there was no effect on growth post weaning. Subsequently the nursery
period was targeted, and the effect of a post weaning starter regime specifically
formulated for LBiW pigs was investigated (Chapter 6).
The main conclusions from this thesis were: BiW has a significant effect on future
growth, with initial differences in BW perpetuated with age. Environmental factors at
different stages of production may limit the growth of LBiW pigs and, as such, these
pigs are disadvantaged by commercial pig production practices. The majority of LBiW
pigs did not compensate for their lower BW as they did not grow significantly faster
than their NBiW pigs counterparts at any later stage. However this thesis did
demonstrate ways of improving growth of LBiW pigs through improved nutrition that
not only allowed them to match the BW of heavier littermates at d 70 of age but that
was also cost effective. Ultimately this thesis has shown that LBiW pigs have the
capacity to grow at the same rate as NBiW pigs and this can be exploited to reduce the
deficit in their relative weight.
90
7.1 Can low birth weight pigs grow as fast as normal birth weight pigs?
In swine literature much emphasis is placed on the poor postnatal growth performance
of LBiW pigs (Rehfeldt and Kuhn, 2006), which may be a consequence of permanent
physiological changes, such as a lower number of muscles fibres (Nissen et al., 2004;
Gondret et al., 2005; Paredes et al., 2013). The data presented in this thesis confirm an
effect of BiW on postnatal growth, with BiW having a significant effect on both ADG
and BW at most stages of production (Chapters 3 to 6). The exception to this was in
finisher pigs, where no effect of BiW was noted (Chapter 5). Despite this, initial
differences in BiW were perpetuated with age; a 500 g average BW advantage of NBiW
pigs at birth, translated to over a 7 kg advantage at slaughter.
It has been stated that LBiW pigs are not able to compensate during postnatal life for
their reduced BW at birth (Rehfeldt and Kuhn, 2006; Wu et al., 2006; Beaulieu et al.,
2010). Whilst the majority of animals studied in this thesis were unable to naturally
meet the BW of heavier littermates without intervention, a small minority were. In
Chapter 3, despite the poor performance associated with pigs of low BiW and WW, a
number of pigs achieved a similar weight for age to higher BiW contemporaries by
finishing. A similar effect was observed in Chapters 4 and 6 where, prior to
experimental diets being fed, a small number of piglets with LBiW were able to meet or
exceed the average weight of NBiW pigs at weaning. These pigs exhibited BW gains
that were similar to or above those of NBiW pigs without any treatments, although this
does not occur in the majority of pigs and what allows some pigs and not others to do
this remains uncertain. Further investigation found no specific traits associated with
these pigs that were able to catch up (e.g. LBiW pigs of a higher BiW or PI) nor were
these piglets from the same birth or cross fostered litter. It could be hypothesized that
the ability to catch up may be a reflection of the intrauterine environment. For example
Handel and Stickland (1988) previously suggested that LBiW pigs that are capable of
catching up with heavier littermates may have a similar number of muscle fibres to
NBiW pigs. However it could just as likely be a result of the postnatal environment, for
example access to superior teats by chance. It would be particularly beneficial to
identify what allows some pigs to catch up with heavier pigs without intervention, as
this could be exploited to help the remainder of LBiW pigs
Whilst the majority of LBiW pigs do not appear to have the capacity to naturally reduce
the deficit in BW, they do have the potential to grow at a similar rate to NBiW pigs
91
during certain stages of production. Both nutritional and management intervention has
demonstrated that growth performance of LBiW pigs can be improved, and as a result,
BW can be increased. (Wolter et al., 2002; Morise et al., 2011; Beaulieu et al., 2012;
Han et al., 2013). However it has not been previously shown that these pigs can achieve
the same BW for age as heavier littermates. The data presented here indicate that, in
fact, these LBiW pigs can reach the same BW as heavier littermates when given
appropriate nutritional treatments, suggesting that commercial pig production practice
is limiting the growth potential of these pigs rather than their just their physiology or
inherent growth potential.
As a result of the database analysis in Chapter 3, LBiW pigs in this thesis were defined
as those which weighed ≤ 1.25 kg. The definition of LBiW pigs is highly variable in the
literature, with what may be classed as LBiW in one paper not necessarily in another.
As it remains uncertain whether BiW has a graded effect (variations in the degree of
damage) or a threshold effect (where piglets below a point will be significantly affected)
on performance, the differences in classification may have an effect on any results
obtained.
7.2 Improving the growth performance of low birth weight pigs
As expected, nutrition played a critical role in the growth performance of LBiW pigs.
During the preweaning period, the piglet is solely reliant on the sow for nutrition and, as
a result of its small size, are at a competitive disadvantage when raised with heavier
littermates which can command better quality teats and stimulate these more vigorously
to achieve favourable partitioning of milk supply (Fraser and Jones, 1975; Cutler et al.,
1999). In agreement with this hypothesis, reducing competition by creating litters of
similar sized pigs improved growth during the latter part of lactation, a time when milk
supply plateaus but piglet demand increases (Klobasa et al., 1987). This resulted in a
500 g advantage at weaning in comparison with LBiW piglets grouped with heavier
littermates. Whilst in Chapter 5 no effect of supplementary milk on growth performance
was noted, the provision of creep feed during lactation may have had an effect on
supplementary milk consumption. If piglets are receiving sufficient additional nutrition
from creep feed then it is possible their supplementary milk intake may decrease.
Therefore, in hindsight, piglets should not have been fed creep feed during lactation and
an improved design would ensure that the only source of nutrition available to piglets is
92
sow milk and supplementary milk. However on a practical basis, the provision of creep
feed is vital as there would be no behavioural or gut conditioning to solid feed (BPEX,
2013a) which is likely to have an impact on post-weaning growth (Kuller et al., 2007).
Post weaning, the piglet is no longer reliant on the sow for nutrition, but instead must
adapt to a solid feed. As a result of their LBiW they are more likely to have a reduced
BW at weaning, which may alter their nutritional requirements. However what these
requirements may be has been cause of some debate. Gluckman et al (2005)
hypothesized that a mismatch of the pre and postnatal environment of an animal could
be detrimental. They state that, as a result of an adverse environment in utero, the foetus
makes irreversible alterations to its development trajectory; as such the animal has
developed under the expectation of a nutritionally deprived environment and are not
adapted for a nutrient rich environment. Subsequently, Nissen et al (2011) investigated
the effect of a low protein diet on the performance of LBiW pigs, although no benefit
was noted. In contrast, it was hypothesized in this thesis and by others (Schinckel et al.,
2003; Beaulieu et al., 2012) that LBiW pigs at weaning may benefit from a diet
increased in nutrient density as well as digestibility. Diets are fed based on the ‘average
pig’ and, as a result, are unlikely to meet the requirements of pigs with BW that deviates
from the normal. These LBiW may not only differ in body composition but also their
digestive function, both of which may affect their dietary requirements. This was
confirmed as the provision of a specialised starter regime demonstrated that LBiW pigs
are not only capable of BW gains which are equal to those of NBiW pigs but that they
can match the BW of the average weight pigs leaving the nursery.
7.2.1 The importance of timing
Despite the improvement of LBiW pigs at weaning with a high specification diet, in
Chapter 4, no difference in performance was observed between LBiW pigs fed either
the standard or high specification feed from 9 to 13 weeks of age. It has been suggested
that different periods of the pig’s life are under different influences and this is one
possible explanation for the differences observed. For example, during the early phase,
from birth to the end of the nursery period, both the environment and BiW may play a
major role in pig performance (Dwyer et al., 1993). This could include the physiological
effects of BiW such as reduced gut maturity at weaning (Michiels et al 2013) and
reduced enzyme activity associated with lower BW (Jensen 1997) which may affect
93
both pre and post weaning performance. However after this period, growth performance
may be partially explained by the number of muscle fibres (Dwyer et al 1993) with
increased growth performance in this period associated with a higher number of muscle
fibres and pigs therefore may be less susceptible to intervention.
However it is also possible that the experimental design may have prevented LBiW pigs
from improving their performance in Chapter 4. Efforts were made to minimise nutrient
intake restriction during the lactation period by the provision of supplementary milk and
grouping with similar sized littermates; as such there was no effect of BiW on ADG
during this period. At weaning, however, all pigs were fed a series of starter diets which
were standard for their age and therefore may not have been adequate for LBiW pigs.
Therefore the experimental design may have inadvertently disadvantaged LBiW piglets
post weaning. This is supported by a significant effect of BiW during the period
between weaning and provision of the experimental diets (d 28 to 63). After this period
from d 63 to 91, there was no significant difference in the performance of NBiW (that
weren’t previously restricted) and LBiW pigs, although LBiW still grew considerably
slower. As the impact of early life experiences and previous growth performance
(Klindt, 2003; Pluske et al., 2005) can affect subsequent growth, it is possible that the
mis-feeding of LBiW pigs affected their future growth. In hindsight, prior to nutritional
treatments being fed, LBiW pigs should have been provided with a diet from weaning
that was specifically formulated for their reduced BW (including increased digestibility
and nutrient specification). This would prevent LBiW piglets from being disadvantaged
and subsequently the actual impact of nutritional treatments applied during the grower
phase could be investigated.
7.2.2 The feed intake and efficiency of low birth weight pigs
As reducing competition from heavier littermates and the provision of a high
specification diet increased growth performance, FI is likely to be partly responsible for
the relationship between BiW and postnatal growth (Dwyer et al., 1993). As such, it is
important to determine whether feed intake is a cause or an effect of reduced ADG in
LBiW pigs (Jones et al., 2012). Whilst many have observed lower FI in LBiW pigs in
comparison to their heavier littermates, that was not necessarily the case in this thesis
and there was not a consistent effect of BiW on feed intake, although it is likely that a
certain amount of this variability is due to the different diets provided. During the post
94
weaning period, in comparison to NBiW pigs, LBiW pigs consumed either the same
amount of feed or less. At no point in the experiments did LBiW pigs’ absolute feed
intake exceed that of NBiW pigs, although their feed intake relative to their BW did, an
effect which has been recognised in recent literature (Krueger et al., 2013).
When considering the difference in FI between LBiW pigs on different nutritional
treatments, there was also no difference in FI. However, pigs fed the high specification
feed had increased ADG. This suggests that FI may be a limiting factor for LBiW piglet
performance (Gondret et al., 2005). In general pigs consume feed to satisfy their energy
requirements (Lammers et al., 2007); however it seems that, whilst LBiW pigs benefit
from a nutrient dense diet, they do not increase their FI when fed a standard diet. One
possible explanation is that physiology of LBiW pigs is preventing them from
increasing their FI. As the gut capacity of weaned pigs may limit feed intake (Varley et
al., 2001), it could be expected that pigs which weigh significantly less at weaning have
a lower capacity than heavier littermates. In addition, the development of the digestive
tract of LBiW piglets is retarded at weaning, with a decreased small intestinal weight:
length ratio (Michiels et al., 2013). Therefore LBiW pigs may already be eating at
capacity, which is why no difference was noted in this thesis when pigs ate diets of
different specifications. If this is the case then it would be expected that, as they mature,
FI would no longer be limiting, which may explain why, in the finisher phase, they were
able to exhibit similar ADG to NBiW. However this would require further investigation.
In addition to morphological limits, it is possible that physiological factors are limiting
the FI of LBiW pigs. Gastric emptying can influence FI in pigs (Gregory et al., 1989)
and as it may be delayed in LBiW humans neonates (Neu, 2007), this may be the same
in LBiW pigs, although there is currently no evidence to support or refute this.
The modern commercial pig is often bred for higher lean tissue growth potential.
Inevitably the composition of the gain will vary amongst individuals and poor
efficiency in animals may be indicative of increased fat and reduced protein deposition
(Hodgson and Coe, 2005). As LBiW pigs may be susceptible to increased fat
deposition, it would be expected that they are less efficient than heavier littermates
(Wolter et al., 2002; Gondret et al., 2006; Bérard et al., 2008; Schinckel et al., 2010).
One theory is that prenatal programming affects the postnatal phenotype of the muscle
(Karunaratne et al., 2005), with LBiW pigs pre-programmed to grow more lipid at the
expense of protein hence their inefficiency. Whilst, at certain points in this thesis, LBiW
95
pigs were less efficient, this was not consistent both within and between experimental
chapters. It is difficult to directly compare the results of Chapters 4 and 6 as pigs were
fed different experimental diets and this is likely to influence FCE. However, in general,
LBiW pigs were either less efficient or similar to NBiW pigs depending on the period in
question. Another theory for reduced feed efficiency in LBiW pigs is that they have
inadequate digestive ability (Gondret et al., 2006; Morise et al., 2008), with normal gut
maturation retarded at birth and at weaning in LBiW pigs (Wang et al., 2005; Michiels
et al., 2013). Although it is not possible to speculate on which ingredients in the high
specification diets used in Chapter 6 improved ADG, the diet was highly digestible,
amongst other things, which may have contributed towards the improved efficiency. It
has also been hypothesized LBiW piglets may have reduced nutrient utilization (Wang
et al., 2008; Jones, 2012) which may be responsible for their decreased efficiency
although as the diets fed in Chapter 6 were both nutrient dense and highly digestible, it
is not possible to separate the two.
In terms of carcass composition, if LBiW pigs were less efficient throughout production
then a carcass with a higher percentage of fat would be expected. Whilst it has been
reported that they have increased fat but lower lean muscle content (Powell and Aberle,
1980; Rehfeldt and Kuhn, 2006), as well as lower tenderness of meat (Gondret et al.,
2006), other studies have found little (Bérard et al., 2008) to no effect (Jones, 2012) of
BiW on carcass composition. These differences in results may arise from differences in
breeds, feeding strategies and variation in the definition of LBiW. However the absence
of a definitive conclusion on the effect of BiW on carcass composition may indicate that
LBiW pigs are not consistently less efficient than heavier littermates.
7.3 What are the long term effects of interventions on growth?
To increase the likelihood of adoption by the Pig Industry, the long term effects of any
successful treatments should be considered. There has been no indication that the
growth trajectory of LBiW pigs has been altered by the treatments applied in this thesis.
Rather, interventions give LBiW pigs a BW advantage which may or may not be
retained in the long term.
A 500 g increase in WW was found with manipulation of litter composition during
lactation; however no subsequent benefit was noted for ADG or BW post weaning.
96
Despite this, a numerical difference was observed at slaughter with LBiW pigs grouped
with similar weight littermates almost 2 kg heavier at slaughter than those in mixed
litters. One possible explanation for this lack of effect post weaning is that, although
differences in BW are maintained to finishing, these relatively small differences cannot
be detected in the latter stages due to the increasing weight variation (Wellock et al.,
2009). Whilst Klindt et al (2003) found that those piglets with heavier BW at weaning
had better post weaning performance, the manner in which a heavier WW is attained
may also be important (Lawlor et al., 2002; Pluske et al., 2005). Piglets with naturally
higher weights are more likely to retain any weight advantage, whereas weight which is
higher as a result of interventions is more likely to be lost in the long term (Lawlor et
al., 2002). It is likely that BiW plays an important part in this observation, as it is
commonly reported that BiW is more important than the effects of any subsequent
treatments (Wolter et al., 2002) and is the greatest determinant of lifetime performance
(Dunshea, 2003). Manipulation of post weaning starter regime also had an effect on
future BW, with a 3 kg improvement at exit from the nursery (Chapter 6). Although this
was not a long term study, LBiW pigs did continue to show improved ADG for 3 weeks
after the experimental feed was consumed. This suggests that there may be long term
effects of post weaning feeding regime on the performance of LBiW pigs. Inevitably,
whether any BW advantage is retained will depend on the future growth of LBiW pigs,
which is likely to be dependent on the environment. If limiting conditions are once
again present, then it is likely that any BW advantage will be lost.
7.4 Management of low birth weight pigs on farm
Effective treatments for improving the performance of LBiW piglets are few and far
between, and, in most cases, variation is managed rather than reduced. The findings
from this thesis have highlighted the importance of the postnatal environment for LBiW
pigs and a number of treatments have been proven successful in improving
performance. It is important to consider not only how these treatments can be applied on
farm, but also whether they are economic. Ultimately, treatments which are
straightforward to implement but also economically viable will have the greatest uptake
by producers.
97
7.4.1 Preweaning management
The results of this thesis suggest that a non-competitive environment is particularly
important to LBiW pigs in early life. Reducing competition from heavier pigs by cross
fostering LBiW piglets to litters with similar weight littermates, improved the WW of
LBiW pigs. In practice, this means cross fostering piglets to create litters which
comprise solely of LBiW pigs. The process of cross fostering, if done correctly, is
unlikely to have detrimental effects on performance (Milligan et al., 2001a; Deen and
Bilkei, 2004; Bishop, 2011). Timing is particularly important; piglets must stay with
sows long enough to consume sufficient colostrum but be moved before a teat hierarchy
can be established (McBride, 1963). It is also beneficial, where possible, to move
heavier littermates and leave the smallest pigs on their birth sow (Barrie, 2006), and to
always ensure the sow has enough functional teats to support the litter. Whilst extensive
cross fostering is likely to be labour intensive, selective cross fostering of heavier
piglets from litters with LBiW piglets is a feasible technique as cross fostering is
undertaken on most EU farms to equalize litter sizes. However it is important to
consider that there may be implications of selective cross fostering for NBiW pigs. If
NBiW pigs are being cross fostered into litters with similar sized littermates this means
that half of these pigs will now get access to the poor teats whereas in a litter with
LBiW pigs they would have had access to the superior anterior teats. This may mean
that in comparison to being grouped with LBiW pigs, they may not grow as well.
Although the provision of supplementary milk did not benefit LBiW piglet performance
or survival, it did reduce BW variation to slaughter weight. Reducing variation would
be favourable on farm, as having a smaller BW range of LBiW pigs would make
management easier. In the majority of literature, supplementary milk has a positive
effect on piglet performance (Azain et al., 1996; Zijlstra et al., 1996; Dunshea et al.,
1998; Dunshea et al., 1999; Wolter et al., 2002), therefore this should not be ruled out as
a management tool for improving piglet performance during lactation. Additionally, if
LBiW piglets are grouped with similar sized littermates, then providing milk to those
litters that need it most will ensure costs are kept low rather than providing milk to a
larger number of litters.
98
7.4.2 Postweaning nutrition
The correct choice of starter feed regime is critical to minimize the post-weaning
growth check as pigs transition from liquid to solid feed. It was demonstrated that
feeding a high specification starter diet with extra feed (corresponding to the last diet of
the starter regime) can not only improve LBiW performance, but result in a similar
nursery exit weight of low BiW and normal BiW pigs (Chapter 6). In practice,
producers will not separate by BiW as done in this thesis, but rather by BW at weaning.
By default, a number of low WW pigs will be low BiW pigs, given the high correlation
between BiW and WW. However, little is known about the difference in the physiology
of pigs which are light at weaning as a result of different factors and therefore how they
respond to treatments. Despite this, it has been shown that pigs which are light at
weaning can benefit from enhanced starter regimes (Mahan and Lepine, 1991;
Magowan et al., 2011b) and, as these studies selected by WW rather than BiW, pigs
were likely to have been of low weight for a number of different causes. It is likely,
therefore, that we can extrapolate results obtained in chapter 6 to low weaning weight
pigs as well.
To be an economically viable treatment, the price of feeding the higher quality diets
must be offset by the gains in BW or improved food utilization (Lawlor et al., 2002).
The results presented here suggest that not only can low birth weight pigs at weaning
benefit from an improved dietary regime, but that it is cost effective for producers with
an increased return per pig, and should be preferred to a standard commercial regime
which has poorer margin over feed cost. In contrast, for NBiW pigs the standard
commercial regime was the least expensive and had the greatest margin over feed cost.
Separation of pigs with low BW at weaning will allow selective feeding of an improved
regime, as heavier pigs are best suited to a standard commercial diet.
7.4.3 The economic implications of low birth weight pigs
It has been suggested that it may be more economical to cull LBiW pigs and some
producers may in fact cull pigs below a certain BiW, on the assumption that they will
not perform well. However, the results presented here confirm that LBiW pigs can
perform very well indeed if managed/fed appropriately. This thesis included some pigs
of particularly low BiW, as low as 0.6 kg, with approximately 20 % of LBiW pigs in
each experiment below 0.8 kg. On closer inspection of particularly LBiW pigs in
99
Chapter 6 (≤ 0.8.kg), the majority of these pigs were able to meet BW of heavier
littermates by d 70. In addition to improved growth rates, it was demonstrated that low
pre weaning mortality rates can be achieved with extra care of LBiW pigs.
Independent of the methods used to improve the performance of LBiW pigs, what is of
particular interest is the trade-off between having increased litter sizes in comparison to
dealing with the problems LBiW pigs may present. The problem faced by producers is
whether to rear LBiW pigs which may or may not perform well, versus culling these
pigs at birth which is a potential loss of earnings. The economic impact of rearing
LBiW pigs will be dependent on their ability to convert feed efficiently, financial
penalties at the abattoir, costs associated with managing variation during production as
well as current pig and feed prices. Therefore the impact of any future increases in litter
size should be weighed against the consequences of LBiW pigs and their potential
impact on system efficiency
7.5 Scope for future research
It was observed in this thesis that a number of pigs with LBiW were able to ‘naturally’
compensate during lactation and the nursery period. In future it would be therefore be
beneficial to identify any common traits at birth amongst these pigs that enable them to
catch up with heavier littermates. It has been suggested that the body shape of piglets at
birth is reflective of foetal growth restriction and has a pronounced influence on early
postnatal development and behaviour (Litten et al., 2001). As demonstrated by Baxter et
al (2008), morphological measurements such as PI can be indicative of survivability in
new born pigs, therefore it is possible that these measurements may also be related to
postnatal performance. Ponderal index may also have a significant influence on the
postnatal changes in levels of plasma IGF-1 from d 3 to 150 in pigs (Litten et al., 2004),
which is important in growth regulation. More recently, head morphology has been
identified as a good indicator of IUGR in pigs and is related to colostrum intake (Amdi
et al., 2013). Therefore morphological characteristics, such as PI, BMI and CRL may be
indicative of the intrauterine environment, and be able to predict postnatal performance
and warrant investigation.
Ultimately prevention of LBiW pigs is preferred to management, although at the present
time this is not possible. Whilst previously prevention has focused on maternal nutrition
100
during gestation, current thinking indicates that causal factors of LBiW pigs occur
earlier than this. For example, increasing the period in between weaning and the next
pregnancy in sows has been shown to reduce the CV and increase the total litter weight
at birth, likely related to enhanced restoration of follicle development (Wientjes et al.,
2013). High ovulation rates have also been associated with intrauterine crowding and
low birth weight pigs (Foxcroft et al., 2006), with a LBiW litter phenotype identified
(Foxcroft et al., 2009). It has been hypothesized that, as a result of intrauterine crowding
driven by high ovulation rates, all pigs in a litter would have this phenotype, in
comparison to non-restricted litters which have a NBiW phenotype. Subsequent work
has confirmed that this LBiW litter phenotype is associated with increased intrauterine
crowding and poor lean growth performance (Smit et al., 2013). Whilst it is early days,
results suggest that genes controlling differences in ovarian follicular development may
contribute to differences in ovulation rate (Foxcroft, 2013), with possible differential
gene expression profile between high and low birth weight litter phenotypes. This
would eventually allow for identification of embryos with high or low potential for
growth and may explain differences in pigs being able to compensate that BiW seems
unable to fully answer.
7.6 Final conclusion and findings from the thesis
The findings of this thesis demonstrate the importance of BiW for future growth and
how initial differences are perpetuated as pigs progress from birth to finishing. Whilst
LBiW pigs may be inherently disadvantaged from birth, the postnatal environment
disadvantages them further, making it difficult for them to catch up. Dietary and
management interventions targeted at specific populations have been used with varying
degrees of success to manipulate future BW.
This thesis also offers novel insights into the crucial stages of postnatal growth for
LBiW pigs, with LBiW pigs benefitting from interventions in earlier stages of
production. Most importantly, it was established that LBiW pigs are able to exhibit
growth rates similar to heavier littermates post weaning, enabling them not only to
decrease the deficit in their weight but also to match the BW of heavier pigs on exit
from the nursery. However, whether a weight advantage is still present at slaughter
remains uncertain. Ultimately this thesis provides much needed practical management
solutions for LBiW pigs that are cost effective.
101
References
Algers, B., and P. Jensen. 1984. Communication during suckling in the domestic pig.
Effects of continuous noise. Applied Animal Behaviour Science 14: 49-61.
Algers, B., and P. Jensen. 1991. Teat stimulation and milk production during early
lactation in sows: effects of continuous noise. Canadian Journal of Animal
Science 71: 51 - 60.
Amdi, C., U. Krogh, C. Flummer, N. Oksbjerg, C. F. Hansen, and P. K. Theil. 2013.
Intrauterine growth restricted piglets defined by their head shape ingest
insufficient amounts of colostrum. Journal of Animal Science 91(12): 5605-5613
AOAC. 1990. Official methods of analysis. 15th ed. Assoc. Off. Anal. Chem.,
Arlington, VA.
Auldist, D. E., L. Morrish, P. Eason, and R. H. King. 1998. The influence of litter size
on milk production of sows. Animal Science 67: 333-337.
Azain, M. J., T. Tomkins, J. S. Sowinski, R. A. Arentson, and D. E. Jewell. 1996. Effect
of supplemental pig milk replacer on litter performance: Seasonal variation in
response. Journal of Animal Science 74: 2195-2202.
Baker, J. E. 2004. Effective environmental temperature. Journal of Swine Health and
Production 12: 130-143.
Baldwin, B. A., and G. B. Meese. 1979. Social behaviour in pigs studied by means of
operant conditioning. Animal Behaviour 27: 947-957.
Barrie, E. 2006. Limited cross-fostering shows results.
http://www.prairieswine.com/limited-cross-fostering-shows-results/ Accessed
January 2014.
102
Bauer, R., B. Walter, P. Brust, F. Fuchtner, and U. Zwiener. 2003. Impact of
asymmetric intrauterine growth restriction on organ function in newborn piglets.
European Journalof Obstetrics Gynecology and Reproductive Biology 110: S40-
S49.
Bauer, R., B. Walter, A. Hoppe, E. Gaser, V. Lampe, E. Kauf, and U. Zwiener. 1998.
Body weight distribution and organ size in newborn swine (sus scrofa
domestica) - A study describing an animal model for asymmetrical intrauterine
growth retardation. Experimental and Toxicologic Pathology 50: 59-65.
Baxter, E. M., S. Jarvis, R. B. D’Eath, D. W. Ross, S. K. Robson, M. Farish, I. M.
Nevison, A. B. Lawrence, and S. A. Edwards. 2008. Investigating the
behavioural and physiological indicators of neonatal survival in pigs.
Theriogenology 69: 773-783.
Baxter, E. M., K. M. D. Rutherford, R. B. D'Eath, G. Arnott, S. P. Turner, P. Sandøe, V.
A. Moustsen, F. Thorup, S. A. Edwards, and A. B. Lawrence. 2013. The welfare
implications of large litter size in the domestic pig II: management factors.
Animal Welfare 22: 219-238.
Beaulieu, A. D., J. L. Aalhus, N. H. Williams, and J. F. Patience. 2010. Impact of piglet
birth weight, birth order, and litter size on subsequent growth performance,
carcass quality, muscle composition, and eating quality of pork. Journal of
Animal Science 88: 2767-2778.
Beaulieu, A. D., J. Shea, and D. Gillis. 2012. Development of diets for low birth-weight
piglets to improve post-weaning growth performance and optimize net returns to
the producer. http://www.prairieswine.com/development-of-diets-for-low-birth-
weight-piglets-to-improve-post-weaning-growth-performance-and-optimize-net-
returns-to-the-producer/ Accessed November 2013.
Bee, G. 2004. Effect of early gestation feeding, birth weight, and gender of progeny on
muscle fiber characteristics of pigs at slaughter. Journal of Animal Science 82:
826-836.
103
Bérard, J., M. Kreuzer, and G. Bee. 2008. Effect of litter size and birth weight on
growth, carcass and pork quality, and their relationship to postmortem
proteolysis. Journal of Animal Science 86: 2357-2368.
Bérard, J., M. Kreuzer, and G. Bee. 2010. In large litters birth weight and gender is
decisive for growth performance but less for carcass and pork quality traits.
Meat Science 86: 845-851.
Berkeveld, M., P. Langendijk, H. M. G. van Beers-Schreurs, A. P. Koets, M. A. M.
Taverne, and J. H. M. Verheijden. 2007. Postweaning growth check in pigs is
markedly reduced by intermittent suckling and extended lactation. Journal of
Animal Science 85: 258-266.
Bilkei, G., and O. Biro. 1999. Der einfluss des geburtsgewichtes auf das absetgewicht,
auf die saugferkelverluste und saugferkelerkrankungen der ferkel. Tierarztl
Umsch 54: 372-377.
Bishop, J. A. 2011. Effects of Cross Fostering in Swine. MSc, North Carolina State
University.
Black, J. L., L. R. Giles, P. C. Wynn, A. G. Knowles, C. A. Kerr, M. R. Jones, A. D.
Strom, N. L. Gallagher, and G. J. Eamens. 2001. A review - Factors limiting the
performance of growing pigs in commercial environments. In: P. D. Cranwell
(ed.) Manipulating Pig Production Viii, Proceedings. Australasian Pig Science
Association (Apsa) - Conference Proceedings No. 8. p 9-36. Australasian Pig
Science Assoc, Werribee.
BPEX. 2013a. Creep feeding. http://smartstore.bpex.org.uk/index.asp?294895 Accessed
January 2014.
BPEX. 2013b. EU Costs of Pig Meat Production http://www.bpex.org.uk/prices-facts-
figures/costings/EUCOPadjustedformonthlyfeedprice.aspx Accessed January
2014.
104
Brumm, M. C., M. Ellis, L. J. Johnston, D. W. Rozeboom, and D. R. Zimmerman. 2002.
Effect of removal and remixing of lightweight pigs on performance to slaughter
weights. Journal of Animal Science 80: 1166-1172.
Campbell, R. G., and A. C. Dunkin. 1982. The effect of birth-weight on the estimated
milk intake, growth and body-composition of sow-reared piglets. Animal
Production 35: 193-197.
Cerisuelo, A., M. D. Baucells, J. Gasa, J. Coma, D. Carrión, N. Chapinal, and R. Sala.
2009. Increased sow nutrition during midgestation affects muscle fiber
development and meat quality, with no consequences on growth performance.
Journal of Animal Science 87: 729-739.
Chellani, H. K., J. Mahajan, A. Batra, S. Suri, N. K. Anand, and S. K. Das. 1990. Fetal
ponderal index in predicting growth retardation. The Indian Journal of Medical
Research 92: 163-166.
Chevaux, E., A. Sacy, Y. Le Treut, and G. Martineau. 2010. Intrauterine Growth
Retardation (IUGR): Morphological and behavioural description. In: IPVS,
Vancouver, Canada. p 209.
Comstock, R. E., L. M. Winters, and J. N. Cummings. 1944. The Effect of Sex on the
Development of the Pig. Journal of Animal Science 3: 120-128.
Cooper, T. A., M. P. Roberts, H. G. Kattesh, and C. J. Kojima. 2009. Effects of
Transport Stress, Sex, and Weaning Weight on Postweaning Performance in
Pigs. Professional Animal Scientist 25: 189-194.
Cutler, R. S., V. A. Fahi, E. M. Spicer, and G. M. Cronin. 1999. Diseases of swine.
Eighth ed. Iowa State University Press, Ames, IA.
D'Inca, R., L. Che, T. Thymann, P. T. Sangild, and I. Le Huërou-Luron. 2010a.
Intrauterine growth restriction reduces intestinal structure and modifies the
response to colostrum in preterm and term piglets. Livestock Science 133: 20-22.
105
D'Inca, R., M. Kloareg, C. Gras-Le Guen, and I. Le Huerou-Luron. 2010b. Intrauterine
Growth Restriction Modifies the Developmental Pattern of Intestinal Structure,
Transcriptomic Profile, and Bacterial Colonization in Neonatal Pigs. Journal of
Nutrition 140: 925-931.
D’Eath, R. B., S. P. Turner, E. Kurt, G. Evans, L. Thölking, H. Looft, K. Wimmers, E.
Murani, R. Klont, A. Foury, S. H. Ison, A. B. Lawrence, and P. Mormède. 2010.
Pigs’ aggressive temperament affects pre-slaughter mixing aggression, stress
and meat quality. Animal 4: 604-616.
De Grau, A. F., C. E. Dewey, R. M. Friendship, and K. De lange. 2005. Observational
study of factors associated with nursery pig performance. Canadian Journal of
Veterinary Research 69: 241-245.
Deen, M. G. H., and G. Bilkei. 2004. Cross fostering of low-birthweight piglets.
Livestock Production Science 90: 279-284.
Devillers, N., J. Le Dividich, and A. Prunier. 2011. Influence of colostrum intake on
piglet survival and immunity. Animal 5: 1605 - 1612.
Dewey, C. E., C. Haley, T. Widowski, Z. Poljak , and R. M. Friendship. 2009. Factors
associated with in-transit losses of fattening pigs. Animal Welfare 18: 355-361.
Douglas, S. L., S. A. Edwards, and I. Kyriazakis. 2014. Management strategies to
improve the performance of low birth weight pigs to weaning and their long-
term consequences. Journal of Animal Science 92: 2280-2288.
Douglas, S. L., S. A. Edwards, E. Sutcliffe, P. W. Knap, and I. Kyriazakis. 2013.
Identification of risk factors associated with poor lifetime growth performance
in pigs. Journal of Animal Science 91: 4123-4132.
Drake, A., D. Fraser, and D. Weary. 2008. Parent–offspring resource allocation in
domestic pigs. Behavioral Ecology and Sociobiology 62: 309-319.
106
Dunshea, F. K., DK; Cranwell, PD; Campbell, RG; Mullan, BP; King, RH; Power, GN;
Pluske, J. R. 2003. Lifetime and post-weaning determinants of performance
indices of pigs. Australian Journal of Agricultural Research 54: 363-370.
Dunshea, F. R., J. M. Boyce, and R. H. King. 1998. Effect of supplemental nutrients on
the growth performance of sucking pigs. Australian Journal of Agricultural
Research 49: 883-888.
Dunshea, F. R., D. J. Kerton, P. J. Eason, and R. H. King. 1999. Supplemental skim
milk before and after weaning improves growth performance of pigs. Australian
Journal of Agricultural Research 50: 1165-1170.
Dunshea, F. R., D. K. Kerton, P. D. Cranwell, R. G. Campbell, B. P. Mullan, R. H.
King, and J. R. Pluske. 2002. Interactions between weaning age, weaning
weight, sex and enzyme supplementation on growth performance of pigs.
Australian Journal of Agricultural Research 53: 939-945.
Dunshea, F. R., and R. J. Van Barneveld. 2003. Colostrum protein isolate enhances gut
development, growth performance and plasma IGF-I and II in neonatal pigs.
Asia Pacific Journal of Clinical Nutrition 12 Suppl: S40.
Dwyer, C. M., J. M. Fletcher, and N. C. Stickland. 1993. Muscle cellularity and
postnatal-growth in the pig. Journal of Animal Science 71: 3339-3343.
Dwyer, C. M., N. C. Stickland, and J. M. Fletcher. 1994. The influence of maternal
nutrition on muscle fiber number development in the porcine fetus and on
subsequent postnatal growth. Journal of Animal Science 72: 911-917.
Edwards, M. 2010. Review of weaning age effects on weaned pig performance London
Swine Conference – Focus on the Future p167-171, London.
Edwards, S. A. 2002. Perinatal mortality in the pig: environmental or physiological
solutions? Livestock Production Science 78: 3-12.
107
English, J. G. H., and G. Bilkei. 2004. The effect of litter size and littermate weight on
pre-weaning performance of low-birth-weight piglets that have been
crossfostered. Animal Science 79: 439-443.
English, P. R. 1998. Ten basic principles of fostering piglets. Pig Progress 14: 39-41.
English, P. R., W. J. Smith, and A. Maclean. 1977. The Sow: Improving her Efficiency.
Farming Press Ltd, Ipswich, UK.
Fix, J. S., J. P. Cassady, W. O. Herring, J. W. Holl, M. S. Culbertson, and M. T. See.
2010. Effect of piglet birth weight on body weight, growth, backfat, and
longissimus muscle area of commercial market swine. Livestock Science 127:
51-59.
Foxcroft, G. R. 2013. Capturing Genetic Merit in Differentiated Pork Production
Systems through Genomics, University of Alberta, PigGen Canada.
Foxcroft, G. R., W. T. Dixon, M. K. Dyck, S. Novak, J. C. S. Harding, and F. R. C. L.
Almeida. 2009. Prenatal programming of postnatal development in the pig.
Control Pig Production 8: 213-231.
Foxcroft, G. R., W. T. Dixon, S. Novak, C. T. Putman, S. C. Town, and M. D. A.
Vinsky. 2006. The biological basis for prenatal programming of postnatal
performance in pigs. Journal of Animal Science 84 Suppl: E105-112.
Fraser, D. 1980. A review of the behavioural mechanism of milk ejection in the
domestic pig. Applied Animal Ethology 6: 247-255.
Fraser, D. 1984. The role of behaviour in swine production: a review of research.
Applied Animal Ethology 11: 317-339.
Fraser, D., and R. M. Jones. 1975. The ‘teat order’ of suckling pigs: I. Relation to birth
weight and subsequent growth. The Journal of Agricultural Science 84: 387-
391.
108
Gluckman, P. D., W. Cutfield, P. Hofman, and M. A. Hanson. 2005. The fetal, neonatal,
and infant environments—the long-term consequences for disease risk. Early
Human Development 81: 51-59.
Gondret, F., L. Lefaucheur, H. Juin, I. Louveau, and B. Lebret. 2006. Low birth weight
is associated with enlarged muscle fiber area and impaired meat tenderness of
the longissimus muscle in pigs. Journal of Animal Science 84: 93-103.
Gondret, F., L. Lefaucheur, L. Louveau, B. Lebret, X. Pichodo, and Y. Le Cozler. 2005.
Influence of piglet birth weight on postnatal growth performance, tissue
lipogenic capacity and muscle histological traits at market weight. Livestock
Production Science 93: 137-146.
Gonyou, H. W., M. C. Brumm, E. Bush, J. Deen, S. A. Edwards, T. Fangman, J. J.
McGlone, M. Meunier-Salaun, R. B. Morrison, H. Spoolder, P. L. Sundberg,
and A. K. Johnson. 2006. Application of broken-line analysis to assess floor
space requirements of nursery and grower-finisher pigs expressed on an
allometric basis. Journal of Animal Science 84: 229-235.
Gonyou, H. W., and C. Peterson. 1998. Pre-sorting pigs by weight. Prairie Swine Centre
Annual Report 1998: 37-39.
Greenwood, P. L., A. W. Bell, P. E. Vercoe, and G. J. Viljoen. 2009. Managing the
Prenatal Environment to Enhance Livestock Productivity. Springer, Dordrecht.
Gregory, P. C., M. McFadyen, and D. V. Rayner. 1989. Relation between gastric
emptying and short-term regulation of food intake in the pig. Physiology &
Behavior 45: 677-683.
Hales, J., V. A. Moustsen, M. B. F. Nielsen, and C. F. Hansen. 2013. Individual
physical characteristics of neonatal piglets affect preweaning survival of piglets
born in a noncrated system. Journal of Animal Science 91: 4991-5003.
Han, F., L. Hu, Y. Xuan, X. Ding, Y. Luo, S. Bai, S. He, K. Zhang, and L. Che. 2013.
Effects of high nutrient intake on the growth performance, intestinal morphology
109
and immune function of neonatal intra-uterine growth-retarded pigs. British
Journal of Nutrition FirstView: 1-9.
Handel, S. E., and N. C. Stickland. 1988. Catch-up growth in pigs- A relationship with
muscle cellularity. Animal Production 47: 291-295.
Hansen, A. V., A. B. Strathe, E. Kebreab, J. France, and P. K. Theil. 2012. Predicting
milk yield and composition in lactating sows: A Bayesian approach. Journal of
Animal Science 90: 2285-2298.
Hazzledine, M. 2008. Premier Atlas: Ingredient Matrix. Premier Nutrition Products
Limited, UK.
Hill, G. M., S. K. Baido, G. L. Cromwell, D. C. Mahan, J. L. Nelssen, H. H. Stein, and
N.-C. o. S. Nutrition. 2007. Evaluation of sex and lysine during the nursery
period. Journal of Animal Science 85: 1453-1458.
Hodgson, D., and C. Coe. 2005. Perinatal Programming. Taylor & Francis.
Hoque, M. A., H. Kadowaki, T. Shibata, T. Oikawa, and K. Suzuki. 2007. Genetic
parameters for measures of the efficiency of gain of boars and the genetic
relationships with its component traits in Duroc pigs. Journal of Animal Science
85: 1873-1879.
Jamin, A., B. Seve, J.-N. Thibault, and N. Floc'h. 2012. Accelerated growth rate
induced by neonatal high-protein milk formula is not supported by increased
tissue protein synthesis in low-birth-weight piglets. Journal of Nutrition and
Metabolism 2012: 545341.
Jones, A. P. 2012. Characterizing the Fallback pig, Iowa State University.
Jones, C. K., N. K. Gabler, R. G. Main, and J. F. Patience. 2012. Characterizing growth
and carcass composition differences in pigs with varying weaning weights and
postweaning performance. Journal of Animal Science 90: 4072-4080.
110
Jones, R. M., R. E. Crump, and S. Hermesch. 2011. Group characteristics influence
growth rate and backfat of commercially raised grower pigs. Animal Production
Science 51: 191-197.
Kapell, D., C. J. Ashworth, P. W. Knap, and R. Roehe. 2011. Genetic parameters for
piglet survival, litter size and birth weight or its variation within litter in sire and
dam lines using Bayesian analysis. Livestock Science 135: 215-224.
Karunaratne, J. F., C. J. Ashton, and N. C. Stickland. 2005. Fetal programming of fat
and collagen in porcine skeletal muscles. Journal of Anatomy 207: 763-768.
Kashyap, S., K. F. Schulze, R. Ramakrishnan, R. B. Dell, and W. C. Heird. 1994.
Evaluation of a mathematical model for predicting the relationship between
protein and energy intakes of low-birth-weight infants and the rate and
composition of weight gain. Pediatric Research 35: 704-712.
Kavanagh, S., P. B. Lynch, P. J. Caffrey, and W. D. Henry. 1997. The effect of pig
weaning weight on post-weaning performance and carcass traits. In:
Manipulating Pig Production VI: Proceedings of the 6th Biennial Conference of
the Australasian Pig Science Association, Canberra, Australian Capital
Territories. p 70-71.
Kerr, J. C., and N. D. Cameron. 1995. Reproductive performance of pigs selected for
components of efficient lean growth. Animal Science 60: 281-290.
Kim, J. H., K. N. Heo, J. Odle, I. K. Han, and R. J. Harrell. 2001. Liquid diets
accelerate the growth of early-weaned pigs and the effects are maintained to
market weight. Journal of Animal Science 79: 427-434.
Kim, J. S., P. L. Shinde, Y. X. Yang, K. Yun, J. Y. Choi, J. D. Lohakare, and B. J.
Chae. 2010. Effects of dietary lactose levels during different starter phases on
the performance of weaning pigs. Livestock Science 131: 175-182.
Kirkwood, R., and G. R. Foxcroft. 2005. Influence of Cross-fostering on Piglet Growth
and Survival.
111
http://www.colostrumresearch.org/Studys/SO57_Influence%20of%20Cross-
fostering%20on%20Piglet%20Growth%20and%20Survival.htm Accessed
January 2013.
Klindt, J. 2003. Influence of litter size and creep feeding on preweaning gain and
influence of preweaning growth on growth to slaughter in barrows. Journal of
Animal Science 81: 2434-2439.
Klobasa, F., E. Werhahn, and J. E. Butler. 1987. Composition of Sow Milk During
Lactation. Journal of Animal Science 64: 1458-1466.
Kościński, K., A. Kozłowska-Rajewicz, M. T. Górecki, M. Kamyczek, and M. Różycki.
2009. Month-of-birth effect on further body size in a pig model. Journal of
Comparative Human Biology 60: 159-183.
Krueger, R., M. Derno, S. Goers, B. Metzler-Zebeli, G. Nuernberg, K. Martens, R.
Pfuhl, C. Nebendahl, A. Zeyner, H. Hammon, and C. Metges. 2013. Higher
body fatness in intrauterine growth retarded juvenile pigs is associated with
lower fat and higher carbohydrate oxidation during ad libitum and restricted
feeding. European Journal of Nutrition 53(2): 583-597.
Kuller, W. I., N. M. Soede, H. M. van Beers-Schreurs, P. Langendijk, M. A. Taverne,
B. Kemp, and J. H. Verheijden. 2007. Effects of intermittent suckling and creep
feed intake on pig performance from birth to slaughter. Journal of Animal
Science 85: 1295-1301.
Kyriazakis, I., and G. C. Emmans. 1991. Diet selection in pigs- dietary choices made by
growing pigs following a period of underfeeding with protein. Animal
Production 52: 337-346.
Kyriazakis, I., and J. G. M. Houdijk. 2007. Food Intake And Performance Of Pigs
During Helath, Disease and Recovery: Paradigms in pig science. Nottingham
University Press.
112
Kyriazakis, I., C. Stamataris, G. C. Emmans, and C. T. Whittemore. 1991. The effects
of food protein-content on the performance of pigs previously given foods with
low or moderate protein contents. Animal Production 52: 165-173.
Lammers, P. J., D. R. Stender, and M. S. Honeyman. 2007. Nutrients for pigs.
http://www.ipic.iastate.edu/publications/310.Nutrients.pdf Accessed January
2014.
Larriestra, A. J., S. Wattanaphansak, E. J. Neumann, J. Bradford, R. B. Morrison, and J.
Deen. 2006. Pig characteristics associated with mortality and light exit weight
for the nursery phase. Canadian Veterinary Journal 47: 560-566.
Lawlor, P. G., P. B. Lynch, P. J. Caffrey, and J. V. O'Doherty. 2002. Effect of pre- and
post-weaning management on subsequent pig performance to slaughter and
carcass quality. Animal Science 75: 245-256.
Lay, D. C., Jr., R. L. Matteri, J. A. Carroll, T. J. Fangman, and T. J. Safranski. 2002.
Preweaning survival in swine. Journal of Animal Science 80: E74-86.
Le Dividich, J. 1999. A review - Neonatal and weaner pig: Management to reduce
variation. In: Manipulating Pig Production Vii, Proceedings, Werribee. p 135-
155.
Le Dividich, J., and P. Herpin. 1994. Effects of climatic conditions on the performance,
metabolism and health status of weaned piglets: a review. Livestock Production
Science 38: 79-90.
Le Dividich, J., M. Vermorel, J. Noblet, J. C. Bouvier, and A. Aumaitre. 1980. Effects
of environmental temperature on heat production, energy retention, protein and
fat gain in early weaned piglets. The British Journal of Nutrition 44: 313-323.
Leibbrandt, V. D., R. C. Ewan, V. C. Speer, and D. R. Zimmerman. 1975. Effect of
weaning and age at weaning on on baby pigs performance. Journal of Animal
Science 40: 1077-1080.
113
Lewis, A. J., V. C. Speer, and D. G. Haught. 1978. Relationship between yield and
composition of sows’ milk and weight gains of nursing pigs. Journal of Animal
Science 47: 634.
Lewis, N. J., and S. Wamnes. 2006. How does weaning weight influence the growth
check in early weaned piglets? http://www.prairieswine.com/pdf/36107.pdf
Accessed January 2014.
Litten, J. C., A. M. Corson, P. C. Drury, and L. Clarke. 2001. The effects of body shape
at birth on the physical and behavioural development of the neonatal pig
Neonatal Society 2001 Summer Meeting, The University of Nottingham.
Litten, J. C., K. S. Perkins, J. Laws, A. M. Corson, and L. Clarke. 2004. Pc42 the Effect
of Ponderal Index on the Plasma Concentration of Insulin-Like Growth Factor-1
(Igf-1) in Neonatal and Growing Pigs. Journal of Pediatric Gastroenterology &
Nutrition 39: S527.
Lynch, P. B., S. Kavanagh, P. G. Lawlor, M. Young, D. Harrington, P. J. Caffrey, and
W. D. Henry. 1998. Effect of pre-and post-weaning nutrition and management
on performance of weaned pigs to circa 35 kg, Teagasc Moorepark Research
Centre.
Magowan, E., M. E. E. Ball, K. J. McCracken, V. E. Beattie, R. Bradford, M. J.
Robinson, M. Scott, F. J. Gordon, and C. S. Mayne. 2011a. Effect of dietary
regime and group structure on pig performance and the variation in weight and
growth rate from weaning to 20 weeks of age. Livestock Science 136: 216-224.
Magowan, E., M. E. E. Ball, K. J. McCracken, V. E. Beattie, R. Bradford, M. J.
Robinson, M. Scott, F. J. Gordon, and C. S. Mayne. 2011b. The performance
response of pigs of different wean weights to 'high' or 'low' input dietary regimes
between weaning and 20 weeks of age. Livestock Science 136: 232-239.
Mahan, D. C. 1992. Efficacy of dried whey and its lactalbumin and lactose components
at two dietary lysine levels on postweaning pig performance and nitrogen
balance. Journal of Animal Science 70: 2182-2187.
114
Mahan, D. C. 1993. Evaluating two sources of dried whey and the effects of replacing
the corn and dried whey component with corn gluten meal and lactose in the
diets of weanling swine. Journal of Animal Science 71: 2860-2866.
Mahan, D. C., G. L. Cromwell, R. C. Ewan, C. R. Hamilton, and J. T. Yen. 1998.
Evaluation of the feeding duration of a phase 1 nursery diet to three-week-old
pigs of two weaning weights. NCR-42 Committee on Swine Nutrition. Journal
of Animal Science 76: 578-583.
Mahan, D. C., N. D. Fastinger, and J. C. Peters. 2004. Effects of diet complexity and
dietary lactose levels during three starter phases on postweaning pig
performance. Journal of Animal Science 82: 2790-2797.
Mahan, D. C., and A. J. Lepine. 1991. Effect of pig weaning weight and associated
nursery feeding programs on subsequent performance to 105 kilograms body-
weight. Journal of Animal Science 69: 1370-1378.
Main, R. G., S. S. Dritz, M. D. Tokach, R. D. Goodband, and J. L. Nelssen. 2004.
Increasing weaning age improves pig performance in a multisite production
system. Journal of Animal Science 82: 1499-1507.
Marchant-Forde, J. N. 2008. The Welfare of Pigs. Springer.
McBride, G. 1963. The “teat order” and communication in young pigs. Animal
Behaviour 11: 53-56.
McCutcheon. 2002. The optimum price for pigs.
http://www.teagasc.ie/publications/2002/pig2002/paper03.asp Accessed January
2014.
McNamara, L. B., L. Giblin, T. Markham, N. C. Stickland, D. P. Berry, J. J. O'Reilly, P.
B. Lynch, J. P. Kerry, and P. G. Lawlor. 2011. Nutritional intervention during
gestation alters growth, body composition and gene expression patterns in
skeletal muscle of pig offspring. Animal 5: 1195-1206.
115
Meade, R. J., J. W. Rust, K. P. Miller, H. E. Hanke, R. S. Grant, L. D. Vermedahl, D. F.
Wass, and L. E. Hanson. 1969. Effects of Protein Level Sequence and Kind of
Starter on Rate and Efficiency of Gain of Growing Swine, and on Carcass
Characteristics. Journal of Animal Science 29: 303-308.
Medel, P., M. A. Latorre, C. de Blas, R. Lázaro, and G. G. Mateos. 2004. Heat
processing of cereals in mash or pellet diets for young pigs. Animal Feed
Science and Technology 113: 127-140.
Menoyo, D., M. P. Serrano, V. Barrios, D. G. Valencia, R. Lázaro, J. Argente, and G.
G. Mateos. 2011. Cereal type and heat processing of the cereal affect nutrient
digestibility and dynamics of serum insulin and ghrelin in weanling pigs.
Journal of Animal Science 89: 2793-2800.
Michiels, J., M. De Vos, J. Missotten, A. Ovyn, S. De Smet, and C. Van Ginneken.
2013. Maturation of digestive function is retarded and plasma antioxidant
capacity lowered in fully weaned low birth weight piglets. British Journal of
Nutrition 109: 65-75.
Miller, Y. J., A. M. Collins, R. J. Smits, P. C. Thomson, and P. K. Holyoake. 2012.
Providing supplemental milk to piglets preweaning improves the growth but not
survival of gilt progeny compared with sow progeny. Journal of Animal Science
90: 5078-5085.
Milligan, B. N., C. E. Dewey, and A. F. de Grau. 2002a. Neonatal-piglet weight
variation and its relation to pre-weaning mortality and weight gain on
commercial farms. Preventive Veterinary Medicine 56: 119-127.
Milligan, B. N., D. Fraser, and D. L. Kramer. 2001a. Birth weight variation in the
domestic pig: effects on offspring survival, weight gain and suckling behaviour.
Applied Animal Behaviour Science 73: 179-191.
116
Milligan, B. N., D. Fraser, and D. L. Kramer. 2001b. The effect of littermate weight on
survival, weight gain, and suckling behavior of low-birth-weight piglets in
cross-fostered litters. Journal of Swine Health and Production 9: 161-166.
Milligan, B. N., D. Fraser, and D. L. Kramer. 2002b. Within-litter birth weight variation
in the domestic pig and its relation to pre-weaning survival, weight gain, and
variation in weaning weights. Livestock Production Science 76: 181-191.
Morise, A., I. Louveau, and I. Le Huerou-Luron. 2008. Growth and development of
adipose tissue and gut and related endocrine status during early growth in the
pig: impact of low birth weight. Animal 2: 73-83.
Morise, A., B. Seve, K. Mace, C. Magliola, I. Le Huerou-Luron, and I. Louveau. 2011.
Growth, body composition and hormonal status of growing pigs exhibiting a
normal or small weight at birth and exposed to a neonatal diet enriched in
proteins. British Journal of Nutrition 105: 1471-1479.
Neu, J. 2007. Gastrointestinal development and meeting the nutritional needs of
premature infants. The American Journal of Clinical Nutrition 85: 629S-634S.
Nielsen, T. T., N. L. Trottier, H. H. Stein, C. Bellaver, and R. A. Easter. 2002. The
effect of litter size and day of lactation on amino acid uptake by the porcine
mammary glands. Journal of Animal Science 80: 2402-2411.
Nienaber, J. A., G. L. Hahn, T. P. McDonald, and R. L. Korthals. 1996. Feeding
patterns and swine performance in hot environments. Transactions of the
American Society of Agricultural Engineers 39: 195-202.
Nissen, P. M., V. O. Danielsen, P. F. Jorgensen, and N. Oksbjerg. 2003. Increased
maternal nutrition of sows has no beneficial effects on muscle fiber number or
postnatal growth and has no impact on the meat quality of the offspring. Journal
of Animal Science 81: 3018-3027.
117
Nissen, P. M., P. F. Jorgensen, and N. Oksbjerg. 2004. Within-litter variation in muscle
fiber characteristics, pig performance, and meat quality traits. Journal of Animal
Science 82: 414-421.
Nissen, P. M., and N. Oksbjerg. 2011. Birth weight and postnatal dietary protein level
affect performance, muscle metabolism and meat quality in pigs. Animal 5:
1382-1389.
O'Connell, M. K., P. B. Lynch, and J. V. O'Doherty. 2006. Determination of the
optimum dietary lysine concentration for boars and gilts penned in pairs and in
groups in the weight range 60 to 100 kg. Animal Science 82: 65-73.
O'Quinn, P. R., S. S. Dritz, R. D. Goodband, M. D. Tokach, J. C. Swanson, J. L.
Nelssen, and R. E. Musser. 2001. Sorting growing-finishing pigs by weight fails
to improve growth performance or weight variation. Journal of Swine Health
and Production 9: 11-16.
Oksbjerg, N., P. M. Nissen, M. Therkildsen, H. S. Møller, L. B. Larsen, M. Andersen,
and J. F. Young. 2013. Meat Science And Muscle Biology Symposium: In utero
nutrition related to fetal development, postnatal performance, and meat quality
of pork. Journal of Animal Science 91: 1443-1453.
Owsley, W. F., D. E. Orr, and L. F. Tribble. 1986. Effects of Age and Diet on the
Development of the Pancreas and the Synthesis and Secretion of Pancreatic
Enzymes in the Young Pig. Journal of Animal Science 63: 497-504.
Pardo, C. E., J. Berard, M. Kreuzer, and G. Bee. 2013a. Intrauterine crowding impairs
formation and growth of secondary myofibers in pigs. Animal 7: 430-438.
Pardo, C. E., M. Kreuzer, and G. Bee. 2013b. Effect of average litter weight in pigs on
growth performance, carcass characteristics and meat quality of the offspring as
depending on birth weight. Animal FirstView: 1-9.
Paredes, S. P., A. J. Jansman, M. W. Verstegen, A. Awati, W. Buist, L. A. den Hartog,
H. M. Van Hees, N. Quiniou, W. H. Hendriks, and W. J. Gerrits. 2012. Analysis
118
of factors to predict piglet body weight at the end of the nursery phase. Journal
of Animal Science 90: 3243-3251.
Paredes, S. P., C. Kalbe, A. J. Jansman, M. W. Verstegen, H. M. van Hees, D. Losel, W.
J. Gerrits, and C. Rehfeldt. 2013. Predicted high-performing piglets exhibit more
and larger skeletal muscle fibers. Journal of Animal Science 91: 5589-5598.
Patience, J. F., and A. D. Beaulieu. 2006. Variation in the Finishing Barn.
http://www.prairieswine.com/pdf/2244.pdf Accessed 2014 January.
Patience, J. F., K. Engele, A. D. Beaulieu, H. W. Gonyou, and R. T. Zijlstra. 2004.
Variation: costs and consequences. In: Advances in pork production:
proceedings of the Banff Pork Seminar. p 257.
Piegorsch, W. W., and J. A. Bailer. 2005. Analysing Environmental Data. John Wiley &
sons, Chichester, West Sussex.
Pluske, J. P., J. Le Dividich, and M. M. W. A. Verstegen. 2003. Weaning the Pig:
Concepts and Consequences. Wageningen Academic Pub.
Pluske, J. R., H. G. Payne, I. H. Williams, and M. B.P. 2005. Early feeding for lifetime
performance of pigs. Recent Advances in Animal Nutrition in Australia No. 15. p
171-181.
Poore, K. R., A. J. Forhead, D. S. Gardner, D. A. Giussani, and A. L. Fowden. 2002.
The effects of birth weight on basal cardiovascular function in pigs at 3 months
of age. The Journal of Physiology 539: 969-978.
Poore, K. R., and A. L. Fowden. 2004. The effects of birth weight and postnatal growth
patterns on fat depth and plasma leptin concentrations in juvenile and adult pigs.
The Journal of Physiology 558: 295-304.
Powell, S. E., and E. D. Aberle. 1980. Effects of birth-weight on growth and carcass
composition of swine. Journal of Animal Science 50: 860-868.
119
Premji, S. S., T. R. Fenton, and R. S. Sauve. 2006. Higher versus lower protein intake in
formula-fed low birth weight infants. The Cochrane database of systematic
reviews: Cd003959.
Puppe, B., J. Lanbein, J. Bauer, and S. Hoy. 2008. A comparative view on social
hierarchy formation at different stages of pig production using sociometric
measures. Livestock Science 113: 155-162.
Quiniou, N., V. Courboulay, Y. Salaün, and P. Chevillion. 2010. Impact of the non
castration of male pigs on growth performance and behaviour-comparison with
barrows and gilts 61st Annual Meeting of the European Association for Animal
Production p1-7. European Heraklion, Crete Island, Greece.
Quiniou, N., J. Dagorn, and D. Gaudré. 2002. Variation of piglets' birth weight and
consequences on subsequent performance. Livestock Production Science 78: 63-
70.
Rehfeldt, C., and G. Kuhn. 2006. Consequences of birth weight for postnatal growth
performance and carcass quality in pigs as related to myogenesis. Journal of
Animal Science 84 Suppl: 113-123.
Rehfeldt, C., L. Lefaucheur, J. Block, B. Stabenow, R. Pfuhl, W. Otten, C. Metges, and
C. Kalbe. 2012. Limited and excess protein intake of pregnant gilts differently
affects body composition and cellularity of skeletal muscle and subcutaneous
adipose tissue of newborn and weanling piglets. European Journal of Nutrition
51: 151-165.
Rehfeldt, C., M. F. W. Te Pas, K. Wimmers, J. M. Brameld, P. M. Nissen, C. Berri, L.
M. P. Valente, D. M. Power, B. Picard, N. C. Stickland, and N. Oksbjerg. 2011.
Advances in research on the prenatal development of skeletal muscle in animals
in relation to the quality of muscle-based food. II – Genetic factors related to
animal performance and advances in methodology. Animal 5: 718-730.
120
Rehfeldt, C., A. Tuchscherer, M. Hartung, and G. Kuhn. 2008. A second look at the
influence of birth weight on carcass and meat quality in pigs. Meat Science 78:
170-175.
Renaudeau, D., M. Kerdoncuff, C. Anaïs, and J. L. Gourdine. 2008. Effect of
temperature level on thermal acclimation in Large White growing pigs. Animal
2: 1619-1626.
Roeder, L. M., and B. F. Chow. 1972. Maternal undernutrition and its long-term effects
on the offspring. American Journal of Clinical Nutrition 25: 812-821.
Roehe, R. 1999. Genetic determination of individual birth weight and its association
with sow productivity traits using Bayesian analyses. Journal of Animal Science
77: 330-343.
Roehe, R., and E. Kalm. 2000. Estimation of genetic and environmental risk factors
associated with pre-weaning mortality in piglets using generalized linear mixed
models. Journal of Animal Science 70: 227-240.
Rothschild, M. F., A. Ruvinsky, and C. A. B. International. 2011. The Genetics of the
Pig. CABI.
Rutherford, K. M. D., E. M. Baxter, R. B. D'Eath, S. P. Turner, G. Arnott, R. Roehe, B.
Ask, P. Sandøe, V. A. Moustsen, F. Thorup, S. A. Edwards, P. Berg, and A. B.
Lawrence. 2013. The welfare implications of large litter size in the domestic pig
I: biological factors. Animal Welfare 22: 199-218.
Schinckel, A. P., M. E. Einstein, S. Jungst, C. Booher, and S. Newman. 2010.
Evaluation of the Impact of Pig Birth Weight on Grow-Finish Performance,
Backfat Depth, and Loin Depth. The Professional Animal Scientist 26: 51-69.
Schinckel, A. P., J. Ferrell, M. E. Einstein, S. A. Pearce, and R. D. Boyd. 2003.
Analysis of Pig Growth From Birth to Sixty Days of Age. The Professional
Animal Scientist 20: 79-86.
121
Skok, J., M. Brus, and D. Škorjanc. 2007. Growth of piglets in relation to milk intake
and anatomical location of mammary glands. Acta Agriculturae Scandinavica,
Section A - Animal Science 57: 129 - 135.
Skorjanc, D., M. Brus, and M. C. Potokar. 2007. Effect of birth weight and sex on pre-
weaning growth rate of piglets. Archiv Fur Tierzucht-Archives of Animal
Breeding 50: 476-486.
Smit, M. N., J. D. Spencer, F. R. C. L. Almeida, J. L. Patterson, H. Chiarini-Garcia, M.
K. Dyck, and G. R. Foxcroft. 2013. Consequences of a low litter birth weight
phenotype for postnatal lean growth performance and neonatal testicular
morphology in the pig. Animal 7: 1681-1689.
Smith, A. L., K. J. Stalder, and T. V. Serenius. 2007. Effect of piglet birth weight on
weights at weaning and 42 days post weaning. Journal of Swine Health and
Production 15: 213-218.
Smith, A. L., K. J. Stalder, T. V. Serenius, T. J. Baas, and J. W. Mabry. 2006. Effect of
weaning age on nursery pig performance, Iowa State University.
Spoolder, H. A., S. A. Edwards, and S. Corning. 2000. Aggression among finished pigs
following mixing in kennelled and unkennelled accommodation. Livestock
Production Science 63: 121-129.
Stamataris, C., I. Kyriazakis, and G. C. Emmans. 1991. The performance and body
composition of young pigs following a period of growth retardation by food
restriction. Animal Production 53: 373-381.
Stanton, H. C., and J. K. Carroll. 1974. Potential mechanisms responsible for prenatal
and perinatal mortality or low viability of swine. Journal of Animal Science 38:
1037-1044.
Stewart, T. S., and M. A. Diekman. 1989. Effect of birth and fraternal litter size and
corss-fostering on growth and reporduction in swine. Journal of Animal Science
67: 635-640.
122
Straw, B. E., J. J. Zimmerman, S. D'Allaire, and D. J. Taylor. 2006. Diseases of Swine
No. 9. p 7. Blackwell Publishing.
Tokach, M. D., R. D. Goodband, J. L. Nelssen, and L. J. Kats. 1992. Influence of
weaning weight and growth during the first week postweaning on subsequent
pig performance Swine Day. Kansas State University. Agricultural Experiment
Station and Cooperative Extension Service Manhattan, KS.
Tuchscherer, M., B. Puppe, A. Tuchscherer, and U. Tiemann. 2000. Early identification
of neonates at risk: traits of newborn piglets with respect to survival.
Theriogenology 54: 371 - 388.
van Wijk, H. J., D. J. G. Arts, J. O. Matthews, M. Webster, B. J. Ducro, and E. F. Knol.
2005. Genetic parameters for carcass composition and pork quality estimated in
a commercial production chain. Journal of Animal Science 83: 324-333.
Varley, M. A., J. Wiseman, and M. British Society of Animal Science. 2001. The
Weaner Pig: Nutrition and Management. CABI Pub.
Wang, J. J., L. X. Chen, D. F. Li, Y. L. Yin, X. Q. Wang, P. Li, L. J. Dangott, W. X.
Hu, and G. Y. Wu. 2008. Intrauterine growth restriction affects the proteomes of
the small intestine, liver, and skeletal muscle in newborn pigs. Journal of
Nutrition 138: 60-66.
Wang, T., Y. J. Huo, F. X. Shi, R. J. Xu, and R. J. Hutz. 2005. Effects of intrauterine
growth retardation on development of the gastrointestinal tract in neonatal pigs.
Biology of the Neonate 88: 66-72.
Wang, T., and R. J. Xu. 1996. Effects of colostrum feeding on intestinal development in
newborn pigs. Biology of the Neonate 70: 339-348.
Wang, X., W. Wu, G. Lin, D. Li, G. Wu, and J. Wang. 2009. Temporal Proteomic
Analysis Reveals Continuous Impairment of Intestinal Development in Neonatal
123
Piglets with Intrauterine Growth Restriction. Journal of Proteome Research 9:
924-935.
Wellock, I. J., G. C. Emmans, and I. Kyriazakis. 2004. Modeling the effects of stressors
on the performance of populations of pigs. Journal of Animal Science 82: 2442-
2450.
Wellock, I. J., J. G. M. Houdijk, A. C. Miller, B. P. Gill, and I. Kyriazakis. 2009. The
effect of weaner diet protein content and diet quality on the long-term
performance of pigs to slaughter. Journal of Animal Science 87: 1261-1269.
Whang, K. Y., F. K. McKeith, S. W. Kim, and R. A. Easter. 2000. Effect of starter
feeding program on growth performance and gains of body components from
weaning to market weight in swine. Journal of Animal Science 78: 2885-2895.
Whittemore, C. T., and I. Kyriazakis (Editors). 2006. Whittemores's Science and
Practice of Pig Production. Blackwell Publishing, 186-195 pp.
Widdowson, E. M. 1971. Intra-uterine growth retardation in the pig. I. Organ size and
cellular development at birth and after growth to maturity. Biology of the
Neonate 19: 329-340.
Wientjes, J. G. M., N. M. Soede, E. F. Knol, H. van den Brand, and B. Kemp. 2013.
Piglet birth weight and litter uniformity: Effects of weaning-to-pregnancy
interval and body condition changes in sows of different parities and crossbred
lines. Journal of Animal Science 91: 2099-2107.
Winder, J. A., J. S. Brinks, R. M. Bourdon, and B. L. Golden. 1990. Genetic analysis of
absolute growth measurements, relative growth rate and restricted selection
indicies in red Angus cattle. Journal of Animal Science 68: 330-336.
Wiseman, J. V., M. A.; Chadwick, J. P. [Editors]. 1998. Progress in Pig Science.
Nottingham University Press.
124
Wolter, B. F., and M. Ellis. 2001. The effects of weaning weight and rate of growth
immediately after weaning on subsequent pig growth performance and carcass
characteristics. Canadian Journal of Animal Science 81: 363-369.
Wolter, B. F., M. Ellis, B. P. Corrigan, and J. M. DeDecker. 2002. The effect of birth
weight and feeding of supplemental milk replacer to piglets during lactation on
preweaning and postweaning growth performance and carcass characteristics.
Journal of Animal Science 80: 301-308.
Wu, G., F. W. Bazer, J. M. Wallace, and T. E. Spencer. 2006. Board-invited review:
Intrauterine growth retardation: Implications for the animal sciences. Journal of
Animal Science 84: 2316-2337.
Wu, G. Y., F. W. Bazer, S. Datta, H. J. Gao, G. A. Johnson, A. Lassala, P. Li, M. C.
Satterfield, and T. E. Spencer. 2008. Intrauterine growth retardation in livestock:
Implications, mechanisms and solutions. Archiv Fur Tierzucht-Archives of
Animal Breeding 51: 4-10.
Xu, R.-J., D. J. Mellor, M. J. Birtles, G. W. Reynolds, and H. V. Simpson. 1994. Impact
of Intrauterine growth Retardation on the Gastrointestinal Tract and the Pancreas
in Newborn Pigs. Journal of Pediatric Gastroenterology and Nutrition 18: 231-
240.
Zijlstra, R. T., K. Y. Whang, R. A. Easter, and J. Odle. 1996. Effect of feeding a milk
replacer to early-weaned pigs on growth, body composition, and small intestinal
morphology, compared with suckled littermates. Journal of Animal Science 74:
2948-2959.
125
126
127